ChatGPT Apps

Find apps that can use eligible ChatGPT plans for AI inference. Account eligibility, app permissions, and usage limits apply. Listings are based on source and documentation inspection; live use is not implied.

About ChatGPT plan usage

Discovery uses GitHub's indexed default-branch code; SDK-only integrations and unindexed code may be missed. Latest search pages were exhausted. Pending stages: 268; overdue repositories: 0. Oldest pending: 2026-10-02.

4 main listings; 219 other discovered repositories. Main-section minimum: 100 stars.

Latest discovery scan #18 (2026-10-03): 26 newly discovered repositories. 72 new or updated repositories. Changes include discoveries, source/release signals, substantive metadata, evidence, and reviews; routine checks and ordinary star changes are excluded. Historical changes use stored discovery, source-reference and review timestamps; full change tracking starts with the next scan.

Available apps / AI agent platform

converge-ai-labs/agent-foundation

Agent Foundation is an open-source, self-hosted agent platform with a Service/Console application and Harness UI. Its released Service supports using an eligible ChatGPT plan through Sign in with ChatGPT, without an OpenAI API key, for agent requests.

AI Usage: The authorized credentials are wired into OpenAIChatGPTProvider, which is passed to OpenAIChatGPTResponsesModel; inference uses the public OpenAI Responses endpoint with the ChatGPT subscription scopes and native streaming profile. This establishes the actual model inference call path rather than a documentation-only mention.

released • release/a13n-service-v0.0.94 • Docker, Web browser, Python 3.13+ • usefulness 8/8 • 100+ stars

refresh pending; assessed 2026-10-02; last successful evidence check 2026-10-02

Details and evidence
  • Install: Use the released Service assets or the release's Compose configuration. The documented deployment starts the Service, Console, PostgreSQL, and Redis with Docker Compose.
  • ChatGPT setup: Start the Service, open the local Console, register the administrator, then add a model under Models and choose Sign in with ChatGPT. The OAuth flow requires a supported callback, renewable credentials, the resource.invoke and chatgpt.tokens.use.direct scopes, and durable Host-managed credential storage.
  • License: Apache License 2.0
  • Requirements: Docker with Docker Compose for the Service deployment; An eligible ChatGPT plan or a model-provider API key; A browser for the local Console; OpenAI approval may be required for hosted or commercial ChatGPT use
  • Usefulness: The evidence covers the integration, inference path, release distribution, installation, setup, requirements, and license.
  • Usefulness: Using an eligible ChatGPT subscription directly for agent requests without an API key is a distinct capability from ordinary API-key provider configuration.
  • Usefulness: The evidence provides a Docker Compose deployment, Console onboarding sequence, model setup step, and OAuth storage and callback requirements.
  • Usefulness: The released Service combines a usable agent platform with direct ChatGPT-plan access, making it practical for users who want self-hosted agent execution without an OpenAI API key.
  • Qualification: The project is in active 0.x development, so APIs and configuration may change between minor releases.
  • Qualification: ChatGPT plan eligibility and model access remain upstream decisions; hosted or commercial use requires separate OpenAI approval.
  • Qualification: The Harness and Service provide protocol and model building blocks but Host applications remain responsible for durable credential storage and authorization.

Evidence: Source 1, Source 2, Source 3, Source 4, Source 5

Upcoming apps / AI assistant

kitknox/rootshell

Rootshell is an open-source Apple-platform terminal app with an apparent Sign in with ChatGPT implementation intended to authorize ChatGPT plan usage and provide OpenAI-backed AI features. The integration is credible but its usability at the assessed commit is unverified because inference-call wiring, packaging, and released setup evidence are missing.

AI Usage: Planned or incomplete integration: Integration is probable rather than confirmed. ChatGPTOAuth.swift implements dynamic client registration, OAuth 2.0 + PKCE/OIDC, renewable sessions, and the chatgpt.tokens.use.direct scope, but no supplied evidence connects the resulting credentials to Rootshell's actual inference call path or SDK/provider wiring.

unknown; upcoming integration • 1cfd8dbdb275 • iPhone, iPad, visionOS, macOS • usefulness 5/8 • 100+ stars; priority watch

refresh pending; assessed 2026-10-02; last successful evidence check 2026-10-02

Details and evidence
  • Development source: The repository documents App Store, TestFlight, direct macOS, and Homebrew distribution for Rootshell, but the supplied evidence does not confirm that the assessed commit's ChatGPT integration is included or enabled in those builds. Source builds can be started by cloning the assessed commit and opening rootshell.xcodeproj in Xcode.
  • Integration proposal: No complete user-facing ChatGPT setup procedure is supplied. The source defines dynamic registration, OAuth 2.0 + PKCE/OIDC, token refresh, and the chatgpt.tokens.use.direct scope, but the evidence does not show the settings UI, provider wiring, or an inference call that consumes the resulting session.
  • License: MIT License
  • Requirements: ChatGPT account with the plan-use scope enabled; Apple platform meeting iOS/iPadOS 18.0+, macOS 14.0+, or visionOS 26.0+; Xcode 26.4+ for building from source; Apple Silicon Mac for iOS Simulator builds
  • Usefulness: The repository documents a broad app and AI feature set, but ChatGPT packaging, release inclusion, and end-to-end call-path evidence are incomplete.
  • Usefulness: The apparent Sign in with ChatGPT flow and direct ChatGPT plan-usage scope are a distinctive capability compared with ordinary API-key-only integrations.
  • Usefulness: The evidence lacks a documented ChatGPT setup flow and does not prove that authorization reaches a usable inference provider.
  • Usefulness: If completed and enabled, ChatGPT plan access alongside terminal AI, SSH, cloud, and Kubernetes capabilities would be highly useful for developers.
  • Qualification: The supplied release listing is empty, and the target metadata has no version or distribution record.
  • Qualification: The ChatGPT source file demonstrates OAuth and plan-scope handling but not the app's inference-provider call path or SDK wiring.
  • Qualification: The documented AI provider list establishes OpenAI support generally, but does not explicitly state that ChatGPT subscription authorization is used for inference.
  • Qualification: Upcoming integration; availability not established

Evidence: Source 1, Source 2, Source 3, Source 4, Source 5, Source 6

Upcoming apps / coding assistant

cline/cline

Cline is an open-source coding agent for IDEs, terminals, and desktop, with a proposed opt-in ChatGPT Plan provider using Sign in with ChatGPT and OpenAI's Responses API. The proposed integration validates the ChatGPT plan grant, discovers account-specific models, and performs stateless streaming inference, but is not released in the assessed targets.

AI Usage: Planned or incomplete integration: Integration is probable rather than confirmed: the open proposal describes a separate opt-in ChatGPT Plan provider, while the assessed unmerged source implements grant validation, account-specific model discovery, and a restricted streaming Responses transport. The evidence does not establish that the provider is wired into a released application or that real account-backed inference has succeeded.

unreleased; upcoming integration • c7f79bf7daae • VS Code, CLI, desktop, macOS, Windows, Linux • usefulness 5/8 • 100+ stars; priority watch

refresh pending; assessed 2026-10-02; last successful evidence check 2026-10-02

Details and evidence
  • Development source: No supported installation route for the ChatGPT Plan provider is established because the implementation target is an unmerged pull request. Do not treat the source tree as a released provider package.
  • Integration proposal: The proposal describes the intended setup through ChatGPT sign-in and provider configuration, but no supported released setup flow is available. The proposal also states that real ChatGPT authorization and plan-backed inference were not performed.
  • License: Apache License 2.0
  • Requirements: A ChatGPT account eligible for the relevant plan access; Sign in with ChatGPT OAuth authorization; The chatgpt.tokens.use.direct grant; Internet access to api.openai.com; Cline CLI, desktop, or VS Code integration containing the provider
  • Usefulness: The proposal covers authorization, token handling, model discovery, inference transport, UI, CLI, desktop, and tests, but real account validation, maintainer review, native UI evidence, and shipped packaging remain unresolved.
  • Usefulness: This is a distinct plan-authenticated ChatGPT provider rather than ordinary OpenAI API-key access or the existing Codex subscription path.
  • Usefulness: The intended OAuth and provider flow is described, but the feature is unmerged and no released installation or verified account setup path is available.
  • Usefulness: The proposed integration would let Cline use account-specific ChatGPT Plan models through an authenticated, restricted Responses API route, giving the coding agent access to plan-backed inference.
  • Qualification: The ChatGPT Plan implementation is in pull request 14689 and is explicitly marked merged:false and state:open.
  • Qualification: The proposal states that no real ChatGPT account authorization or billed/plan-backed inference was performed.
  • Qualification: The released SDK and desktop targets do not provide supplied evidence that this provider was packaged or enabled.
  • Qualification: Upcoming integration; availability not established

Evidence: Source 1, Source 2, Source 3, Source 4, Source 5, Source 6

Upcoming apps / voice translation

kapitalismho/PuriPuly-heart

PuriPuly is a Windows real-time two-way voice translator for VRChat, Discord, and similar environments.

AI Usage: ChatGPT appears intended to provide plan-authorized translation capability after OAuth consent, but the actual ChatGPT inference task and request path are not established by the supplied source. The demonstrated translation implementation instead targets local OpenAI-compatible endpoints.

unreleased; upcoming integration • 991ef0bdf06d • Windows • usefulness 4/8 • 100+ stars; priority watch

current; assessed 2026-10-03

Details and evidence
  • Development source: No supported install route for the ChatGPT integration is established at this source-only target; use the pinned source only for inspection or development.
  • Integration proposal: The documented application setup covers the general translator, Discord authentication, and other model providers, but does not document enabling a ChatGPT plan integration.
  • License: GNU Affero General Public License v3.0 or later
  • Requirements: A ChatGPT account with the required plan scope, if the unreleased integration is completed; Windows for the desktop application; Audio input and VRChat OSC for VRChat subtitle/control features; External speech or language-model services may require separate credentials; local models are also supported
  • Usefulness: The app, source target, license, and general setup are documented, but the ChatGPT inference and distribution chain is incomplete.
  • Usefulness: A plan-authorized ChatGPT path would be distinct from the app's documented local and OpenAI-compatible provider support.
  • Usefulness: No setup instructions establish how to activate or use the ChatGPT integration.
  • Usefulness: Real-time contextual voice translation is useful, but ChatGPT-specific usefulness remains unverified without an inference provider path.
  • Qualification: The documented shipped translation path uses local models and OpenAI-compatible APIs rather than demonstrating ChatGPT inference.
  • Qualification: The ChatGPT OAuth file and session file are present only at the source-only target; the supplied v2.7.0 and v2.6.1 evidence instead shows missing ChatGPT OAuth at those release commits.
  • Qualification: Upcoming integration; availability not established

Evidence: Source 1, Source 2, Source 3, Source 4, Source 5, Source 6, Source 7, Source 8, Source 9

Other discovered repositories

Public source matches below the star threshold, awaiting assessment, or not meeting recommendation criteria. These are discovery records, not confirmed app recommendations. Descriptions may come from unverified GitHub metadata; missing information is unknown.

AdilShamim8/ReWoo

AI teammates that finish the work, and only ever use the parts of your life you allow.

AI Usage: Unknown; assessment pending.

0 stars

Status and source references
  • Below the main-section minimum: 0 stars; 100 required.
  • Assessment pending; source references alone do not establish an app or integration.

Source references: engines/openclaw/extensions/openai/token-sharing-oauth.runtime.test.ts, engines/openclaw/extensions/openai/token-sharing-oauth.runtime.ts, engines/openclaw/extensions/openai/token-sharing.ts

Android-XR-Bridge/AXRB

AXRB is a Windows launcher and compatibility runtime for running owned Android XR and Quest applications through OpenXR.

AI Usage: ChatGPT is intended to analyze consented, bounded, redacted launcher evidence and return streaming diagnostic guidance, optionally using web search. The implementation requests ChatGPT plan usage through OAuth scopes and calls the Responses API, but packaged-user activation is not established.

170 stars

Status and source references
  • availability: contradicted. The literal source_available value requires a runnable ChatGPT plan integration at the target. The evidence shows source-level modules and tests, but not launcher wiring or inclusion and enablement in the released package; the release evidence only lists generic installer and source assets.
  • chain:authorization: unresolved. OAuth authorization, identity validation, and plan-scope enforcement are implemented and tested, but the evidence does not establish that the launcher exposes this authorization flow or connects it to an end-user packaged path. The authorization-to-inference chain is therefore not usablely established.
  • chain:inference: unresolved. The diagnostics class is tested with an injected access provider and a mocked fetch implementation. The evidence proves request behavior but does not prove runnable production wiring from ChatGPTAuth into launcher inference.
  • Missing required evidence: Production launcher wiring from the ChatGPTAuth and AiDiagnostics modules to an end-user UI or command path.
  • Missing required evidence: Documentation showing how packaged users consent to ChatGPT plan usage and start an AI diagnostics request.
  • Missing required evidence: Packaging evidence confirming that the ChatGPT integration is included and enabled in the v1.0.3 Windows installer.
  • Confidence: medium. The target contains ChatGPT OAuth, plan-scope validation, encrypted credential storage, and an AI diagnostics request path with tests, but the supplied evidence does not show the launcher UI or production wiring that enables the feature for packaged users.
  • Open-source licensing is not established.

Source references: launcher/core/chatgpt-auth.mjs, launcher/tests/ai-diagnostics.test.mjs

Anemos-labs/Sleipnir

Unknown; no repository description or verified summary is available.

AI Usage: Unknown; assessment pending.

0 stars

Status and source references
  • Below the main-section minimum: 0 stars; 100 required.
  • Assessment pending; source references alone do not establish an app or integration.

Source references: docs/PROVIDERS.md, internal/chatgptauth/chatgptauth.go

AojdevStudio/nzube

Your intent, ready for your agent. Turns rough ideas into agent-ready briefs, grounded in your code and context. Open source. Available on desktop and mobile.

AI Usage: Unknown; assessment pending.

0 stars

Status and source references
  • Below the main-section minimum: 0 stars; 100 required.
  • Assessment pending; source references alone do not establish an app or integration.

Source references: tools/subscription-proof/README.md, tools/subscription-proof/RECEIPT.md, tools/subscription-proof/src/cli.ts, tools/subscription-proof/src/credentials.ts, tools/subscription-proof/test/mock-openai.ts

Bino5150/lumina

A full featured, powerful, and efficient AI Agentic Harness/Desktop Agent app designed from the ground up with local inference on consumer hardware in mind. It evolves, grows, and gets smarter as you go. And it REMEMBERS...

AI Usage: Unknown; assessment pending.

181 stars

Status and source references
  • Assessment pending; source references alone do not establish an app or integration.

Source references: core/chatgpt_auth/oauth.py, core/chatgpt_auth/session.py, core/chatgpt_auth/store.py, tests/chatgpt_auth_fakes.py, tests/test_chatgpt_auth_01c_protocol.py

Blue-B/pi-custom-packages

Ten extensions for the pi coding agent: session guards, image context control, and Windows/WSL screen capture, cursor, and recording.

AI Usage: Unknown; assessment pending.

1 stars

Status and source references
  • Below the main-section minimum: 1 stars; 100 required.
  • Assessment pending; source references alone do not establish an app or integration.

Source references: packages/pi-openai-accounts/tests/openai-oauth-sdk.test.ts

CaT-Hode/ASS-Agent-Switching-Servant

ASS，让模型切换更顺手。Windows 桌面 AI 路由与账户助手，支持多客户端、官方授权、逐模型检测与自动检查更新。

AI Usage: Unknown; assessment pending.

0 stars

Status and source references
  • Below the main-section minimum: 0 stars; 100 required.
  • Assessment pending; source references alone do not establish an app or integration.

Source references: core/oauth-import.cjs, docs/OAUTH-COMPATIBILITY.md, tests/oauth-qa.test.cjs, tests/oauth-transfer.test.cjs

Community-Access/quill

QUILL Lite is a screen-reader-focused Windows text editor with ChatGPT-plan AI assistance for writing tools, document questions, image description, conversations, and optional web search. The integration is released in QUILL Lite 1.1.2, but the supplied source evidence does not expose the final inference-call wiring.

AI Usage: Planned or incomplete integration: The integration is probable rather than fully confirmed: the release explicitly describes using the user's ChatGPT subscription, and the target source implements OAuth/PKCE consent, plan scope validation, refresh-token storage, and access-token refresh, but the supplied files do not show the downstream inference client or UI-to-inference call path.

45 stars

Status and source references
  • Below the main-section minimum: 45 stars; 100 required.
  • availability: unresolved. A non-prerelease QUILL Lite 1.1.2 release with installer and portable assets exists, but confirmed available-app status requires a complete usable authorization-to-inference chain, and the supplied evidence does not show the inference client or UI call path.
  • chain:inference: unresolved. The source exposes access-token acquisition and refresh but does not identify an inference SDK/client, construct an inference request, or connect the token to the claimed writing, document, image, conversation, or web-search features.
  • Missing required evidence: The exact QUILL Lite UI-to-inference call path is not included in the supplied evidence.
  • Missing required evidence: The specific inference SDK/client implementation and request construction are not included in the supplied evidence.
  • Confidence: medium. The shipped QUILL Lite 1.1.2 release documents ChatGPT-plan use, and the assessed commit contains the complete OAuth/PKCE consent and token-management implementation. However, the supplied evidence does not show the downstream inference client or the UI-to-inference call path, so end-to-end wiring is not fully established.

Source references: quill/core/ai/chatgpt_account.py

CrowBe/weave

Agent Harness for System One model

AI Usage: Unknown; assessment pending.

0 stars

Status and source references
  • Below the main-section minimum: 0 stars; 100 required.
  • Assessment pending; source references alone do not establish an app or integration.

Source references: Unknown; no source file or discussion collected yet.

Cyb3rb1ade/PLUR1BUS-Harness

PLUR1BUS Harness — self-hosted multi-agent harness grown around the PLUR1BUS memory core (TypeScript, Node 24+)

AI Usage: Unknown; assessment pending.

0 stars

Status and source references
  • Below the main-section minimum: 0 stars; 100 required.
  • Assessment pending; source references alone do not establish an app or integration.

Source references: docs/adr/ADR-005-auth-policy-and-secrets.md, docs/milestones.md, docs/provider-matrix.md, docs/superpowers/specs/2026-09-30-openai-auth-design.md

Demonbane18/relmio

Relmio is an Apache-2.0 local and hosted app that connects ChatGPT/Codex sign-in to local apps and self-hosted n8n. Its hosted chat sends request-bound credentials to a server route that forwards prompts and streams completions; its local workflow can copy complete Codex credentials into a private n8n sidecar after confirmation. The OpenAI bridge is unofficial and policy-uncertain, not the documented Sign in with ChatGPT plan-sharing integration.

AI Usage: The hosted inference call path is browser sign-in, request-bound authorization headers, server route credential extraction, and AI SDK streaming. The local n8n path copies the complete credential JSON into the selected private sidecar, where n8n sends supported prompts through the pinned OAuth package. Both paths are documented at the assessed stable target, though no live request was performed.

56 stars

Status and source references
  • Below the main-section minimum: 56 stars; 100 required.
  • chain:authorization: unresolved. User sign-in and explicit local/VPS credential-copy consent are wired, but the evidence does not establish Relmio-specific provider approval, actual granted plan-usage scopes, application registration, or authorization for the credential-copy bridge.
  • Missing required evidence: No live sign-in or inference test was performed.
  • Missing required evidence: No evidence confirms Relmio-specific OpenAI registration, partner approval, or plan-usage grant.
  • Missing required evidence: No deployed-hosting acceptance verifies that the released hosted service matches the assessed source.
  • Missing required evidence: No independent verification establishes provider or intermediary retention and logging behavior.
  • Confidence: medium. Relmio documents and implements a ChatGPT/Codex sign-in-to-inference path, including browser sign-in, request-bound credentials, an AI SDK server route, and streamed responses. However, the hosted flow is explicitly unofficial and policy-uncertain, provider approval and granted plan-usage scopes are unverified, and no live inference or deployed-byte acceptance was performed.

Source references: docs/openai-source-check-2026-10-01.md

ExpressGradient/lazarus

Lazarus is a persistent-IPython coding agent that can use eligible ChatGPT plans for repository inspection, editing, command execution, and long-running coding tasks.

AI Usage: ChatGPT provides the model reasoning and tool selection, while Lazarus exposes persistent Python, job-control, image, and context-loop tools for coding workflows. The provider uses Sign in with ChatGPT OAuth and sends Responses API inference through the authorized plan allowance.

2 stars

Status and source references
  • Below the main-section minimum: 2 stars; 100 required.

Source references: src/lazarus/chatgpt_auth.py

Fallstop/codeslop

my custom slop harness shamelessly stolen from the nearest opensource dev

AI Usage: Unknown; assessment pending.

0 stars

Status and source references
  • Below the main-section minimum: 0 stars; 100 required.
  • Assessment pending; source references alone do not establish an app or integration.

Source references: apps/server/src/provider/CodexChatGptAuth.test.ts, apps/server/src/provider/CodexChatGptAuth.ts, apps/server/src/provider/CodexManagedRuntime.test.ts, apps/server/src/provider/Drivers/CodexDriver.test.ts, apps/server/src/provider/Drivers/CodexManagedProvider.ts

Francigofree/r2b4

Unknown; no repository description or verified summary is available.

AI Usage: Unknown; assessment pending.

1 stars

Status and source references
  • Below the main-section minimum: 1 stars; 100 required.
  • Assessment pending; source references alone do not establish an app or integration.

Source references: integralando/payload/r2b4_voice/openai_oauth.py

IceblueSakura/OpenBridge

Unknown; no repository description or verified summary is available.

AI Usage: Unknown; assessment pending.

2 stars

Status and source references
  • Below the main-section minimum: 2 stars; 100 required.
  • Assessment pending; source references alone do not establish an app or integration.

Source references: docs/references/chatgpt-login.md

Joaov41/Rss

Unknown; no repository description or verified summary is available.

AI Usage: Unknown; assessment pending.

16 stars

Status and source references
  • Below the main-section minimum: 16 stars; 100 required.
  • Assessment pending; source references alone do not establish an app or integration.

Source references: RSSReaderApp/Services/ChatGPTPlanService.swift

Kudl1k/AudioNotes

Unknown; no repository description or verified summary is available.

AI Usage: Unknown; assessment pending.

0 stars

Status and source references
  • Below the main-section minimum: 0 stars; 100 required.
  • Assessment pending; source references alone do not establish an app or integration.

Source references: AudioNotes/Services/ChatGPT/ChatGPTAccount.swift, AudioNotes/Services/ChatGPT/ChatGPTAuthService.swift, AudioNotesTests/ChatGPTAuthTests.swift, AudioNotesTests/ProviderSettingsTests.swift

KvFxKaido/Push

Run your repo from a chat — the git tool that feels like your AI app. Collapses GitHub, terminal, CI, code, and AI into one conversation. Mobile-first, self-hosted, provider-flexible.

AI Usage: Unknown; assessment pending.

3 stars

Status and source references
  • Below the main-section minimum: 3 stars; 100 required.
  • Assessment pending; source references alone do not establish an app or integration.

Source references: docs/decisions/Sign in with ChatGPT — Provider OAuth Assessment.md

KyrosHendrix/pi-KillerOS

TUI, goals, and workflow automation for the Pi coding agent

AI Usage: Unknown; assessment pending.

2 stars

Status and source references
  • Below the main-section minimum: 2 stars; 100 required.
  • Assessment pending; source references alone do not establish an app or integration.

Source references: test/PiExtensionContract.test.ts

MOLAaaaaaaa/MolaGPT.Desktop

Native Windows desktop client for MolaGPT

AI Usage: Unknown; assessment pending.

15 stars

Status and source references
  • Below the main-section minimum: 15 stars; 100 required.
  • Assessment pending; source references alone do not establish an app or integration.

Source references: src/MolaGPT.Core/Auth/ChatGptAccount.cs

MYLEGND/masterapp

MasterApp

AI Usage: Unknown; assessment pending.

0 stars

Status and source references
  • Below the main-section minimum: 0 stars; 100 required.
  • Assessment pending; source references alone do not establish an app or integration.

Source references: AgentPortal.Tests/FounderEngineeringActivationContractTests.cs, AgentPortal.Tests/LegendEngineeringControlPlaneTests.cs, AgentPortal/Services/Engineering/ChatGptPlanResponsesAdapter.cs, AgentPortal/Services/Engineering/LegendChatGptPlanCredentialAuthority.cs

ManoloRemiddi/augmentor-agent

Augmentor Agent Desktop and Browser. Public source under MIT with Augmentor Resale Restriction: free use and modification; resale requires permission.

AI Usage: Unknown; assessment pending.

19 stars

Status and source references
  • Below the main-section minimum: 19 stars; 100 required.
  • Assessment pending; source references alone do not establish an app or integration.

Source references: packages/codex-runtime/src/chatgpt-accounts.ts, packages/codex-runtime/src/chatgpt-auth.ts, scripts/proof-codex-credentials.mjs, tests/codex-chatgpt-accounts.test.mjs, tests/codex-chatgpt-auth.test.mjs

MarcoDotIO/OpenClawKit

OpenClawKit is a Swift SDK for embedding OpenClaw-style agents and running them with a user's ChatGPT plan.

AI Usage: After the user completes Sign in with ChatGPT and grants plan usage, the SDK obtains and refreshes a plan access token, then routes model generation through the ChatGPT Responses API with streaming, tools, reasoning, and multimodal inputs. The SDK exposes this through ChatGPTPlanModelProvider and the OpenAI Responses engine.

26 stars

Status and source references
  • Below the main-section minimum: 26 stars; 100 required.

Source references: Sources/OpenClawCore/SignInWithChatGPT/SignInWithChatGPTConfiguration.swift, Sources/OpenClawCore/SignInWithChatGPT/SignInWithChatGPTErrors.swift, Tests/OpenClawLinuxRuntimeTests/SignInWithChatGPTAuthorizationTests.swift, Tests/OpenClawLinuxRuntimeTests/SignInWithChatGPTSessionTests.swift, Tests/OpenClawLinuxRuntimeTests/SignInWithChatGPTTestSupport.swift

Marve10s/simple-t3code

SimpleT3Code is a cross-platform desktop coding-agent interface that connects Codex to ChatGPT accounts and other locally installed agents.

AI Usage: ChatGPT provides the authenticated Codex model access used for coding-agent tasks, including text generation and model discovery. The app requests ChatGPT token-sharing consent and passes the resulting authenticated runtime into Codex operations.

1 stars

Status and source references
  • Below the main-section minimum: 1 stars; 100 required.

Source references: apps/server/src/provider/CodexChatGptAuth.ts, apps/server/src/provider/Drivers/CodexManagedProvider.ts

Nana1873/stardew-i18n-translator

Portable Windows desktop app for translating Stardew Valley and SMAPI mod i18n files.

AI Usage: Unknown; assessment pending.

0 stars

Status and source references
  • Below the main-section minimum: 0 stars; 100 required.
  • Assessment pending; source references alone do not establish an app or integration.

Source references: src-tauri/src/chatgpt_auth.rs

NateWeav/t3code-hermes

T3 Code fork adding Hermes Agent as an ACP-backed provider

AI Usage: Unknown; assessment pending.

0 stars

Status and source references
  • Below the main-section minimum: 0 stars; 100 required.
  • Assessment pending; source references alone do not establish an app or integration.

Source references: apps/server/src/provider/CodexChatGptAuth.test.ts, apps/server/src/provider/CodexChatGptAuth.ts, apps/server/src/provider/CodexManagedRuntime.test.ts, apps/server/src/provider/Drivers/CodexDriver.test.ts, apps/server/src/provider/Drivers/CodexManagedProvider.ts

Nitestack/aix

Profile-aware credential injector for AI tools backed by a LiteLLM-compatible AI gateway

AI Usage: Unknown; assessment pending.

3 stars

Status and source references
  • Below the main-section minimum: 3 stars; 100 required.
  • Assessment pending; source references alone do not establish an app or integration.

Source references: Unknown; no source file or discussion collected yet.

NousResearch/hermes-agent

The agent that grows with you

AI Usage: Unknown; assessment pending.

250776 stars

Status and source references
  • Assessment pending; source references alone do not establish an app or integration.

Source references: Unknown; no source file or discussion collected yet.

OrdalieTech/orb

Faithful, slim, pure-Go port of the pi coding agent

AI Usage: Unknown; assessment pending.

4 stars

Status and source references
  • Below the main-section minimum: 4 stars; 100 required.
  • Assessment pending; source references alone do not establish an app or integration.

Source references: ai/auth/oauth/openai_chatgpt.go

OutThisLife/dearcv

DearCV is a web app that uses AI to create and edit visually faithful PDF resumes from uploads, websites, GitHub profiles, and chat instructions.

AI Usage: ChatGPT can read resume PDFs and other supplied context, answer in chat, search the web where supported, and edit the open resume through namespaced function tools. The app connects those requests to the signed-in user's ChatGPT plan through a server-held token and OpenAI Responses API integration.

1 stars

Status and source references
  • Below the main-section minimum: 1 stars; 100 required.
  • Open-source licensing is not established.

Source references: lib/chatgpt.ts

Qin-long/sub2api

基于 Sub2API 由多位站长持续维护的独立分支,更加适合中转站长使用，高频更新+同步上游，扩展实用功能并发布生产版本。

AI Usage: Unknown; assessment pending.

0 stars

Status and source references
  • Below the main-section minimum: 0 stars; 100 required.
  • Assessment pending; source references alone do not establish an app or integration.

Source references: Unknown; no source file or discussion collected yet.

Rasalas/tau

A workbench for coding agents: Pi, Claude Code, Codex, OpenCode and Gemini in one window, extensible through kits.

AI Usage: Unknown; assessment pending.

0 stars

Status and source references
  • Below the main-section minimum: 0 stars; 100 required.
  • Assessment pending; source references alone do not establish an app or integration.

Source references: kits/codex/chatgpt-plan-oauth.ts, kits/codex/host.test.ts

Saikrishna-Ambeti/kairo

Unknown; no repository description or verified summary is available.

AI Usage: Unknown; assessment pending.

3 stars

Status and source references
  • Below the main-section minimum: 3 stars; 100 required.
  • Assessment pending; source references alone do not establish an app or integration.

Source references: apps/server/src/provider/CodexChatGptAuth.test.ts, apps/server/src/provider/CodexChatGptAuth.ts, apps/server/src/provider/CodexManagedRuntime.test.ts, apps/server/src/provider/Drivers/CodexDriver.test.ts, apps/server/src/provider/Drivers/CodexManagedProvider.ts

SergeSerb2/t3-pretty

A personal T3 Code fork that keeps every upstream feature while adding T3 Pretty branding, themes, and scenery.

AI Usage: Unknown; assessment pending.

0 stars

Status and source references
  • Below the main-section minimum: 0 stars; 100 required.
  • Assessment pending; source references alone do not establish an app or integration.

Source references: apps/server/src/provider/CodexChatGptAuth.test.ts, apps/server/src/provider/CodexChatGptAuth.ts, apps/server/src/provider/CodexManagedRuntime.test.ts, apps/server/src/provider/Drivers/CodexDriver.test.ts, apps/server/src/provider/Drivers/CodexManagedProvider.ts

StevenBuglione/pwacloud

PWACloud is a mobile PWA and plugin host with a locally run personal ChatGPT runtime and signed plugin packages.

AI Usage: The intended ChatGPT usage is authenticated plan-based Responses streaming with account-specific model catalogues, allowing plugins or the personal runtime to send text prompts and receive streamed results. The supplied evidence does not establish the complete production call path, so practical ChatGPT use remains unknown.

0 stars

Status and source references
  • Below the main-section minimum: 0 stars; 100 required.
  • chain:authorization: unresolved. Authorization requirements and PKCE/state/nonce behavior are documented, and the request shaper checks the direct-plan scope, but no supplied evidence connects authorization and callback credentials to the actual provider call.
  • chain:inference: unresolved. The request shape and terminal-event handling are implemented, but the reference module explicitly says it is not a completed provider adapter, so a usable inference path is not established.
  • chain:distribution: unresolved. Signed plugin distributions are evidenced, but the release notes expressly state that signing does not establish model entitlement or completion of ChatGPT acceptance, and no evidence proves that the released artifacts include and enable the ChatGPT plan integration.
  • Missing required evidence: The supplied evidence does not show the SDK wiring from PKCE/state/nonce authorization and callback-issued client IDs to the provider inference call.
  • Missing required evidence: The supplied evidence does not show a completed ChatGPT provider adapter invoking the shaped request with the authorized account credentials.
  • Missing required evidence: The supplied evidence does not establish that a released plugin artifact includes and enables the ChatGPT plan integration.
  • Usefulness 4/8; available-app minimum is 5/8.

Source references: apps/personal-runtime/src/index.ts, packages/provider-chatgpt/src/index.ts, packages/provider-chatgpt/src/oauth.ts, src/reference/ai.ts, tests/e2e/provider-boundary.spec.ts

TamynatorSama/JobPrep

Unknown; no repository description or verified summary is available.

AI Usage: Unknown; assessment pending.

0 stars

Status and source references
  • Below the main-section minimum: 0 stars; 100 required.
  • Assessment pending; source references alone do not establish an app or integration.

Source references: CLAUDE.md, src-tauri/src/chatgpt_auth.rs

TanStack/ai

🤖 Type-safe, provider-agnostic TypeScript AI SDK for streaming chat, tool calling, agents, and multimodal apps across OpenAI, Anthropic, Gemini, React, Vue, Svelte, and Solid.

AI Usage: Unknown; assessment pending.

3160 stars

Status and source references
  • Assessment pending; source references alone do not establish an app or integration.

Source references: packages/ai-openai/src/siwc.ts, packages/ai-openai/tests/siwc.test.ts

Taskov1ch/mecha-maiden

Unknown; no repository description or verified summary is available.

AI Usage: Unknown; assessment pending.

0 stars

Status and source references
  • Below the main-section minimum: 0 stars; 100 required.
  • Assessment pending; source references alone do not establish an app or integration.

Source references: src/mecha_maiden/auth/chatgpt.py

Techopolis/start-testing-sdk

An open-source SDK embeds privacy-conscious issue reporting and optional ChatGPT-assisted report drafting into applications.

AI Usage: ChatGPT can turn a reviewed, redacted incident excerpt into editable issue prose and can monitor selected application logs to suggest drafts; it cannot submit reports or choose issue metadata. The SDK connects this flow to ChatGPT plan inference through OAuth authorization requiring `chatgpt.tokens.use.direct`, then calls the Responses API with the authorized access token.

1 stars

Status and source references
  • Below the main-section minimum: 1 stars; 100 required.

Source references: apple/Sources/StartTestingChatGPT/ChatGPTAuthorization.swift, apple/Sources/StartTestingChatGPT/ChatGPTTransport.swift, apple/Tests/StartTestingTests/ChatGPTTests.swift, apple/Tests/StartTestingTests/SDKTests.swift, docs/chatgpt.md

UnknownAlienHuman/eliot-swarm-controller

Unknown; no repository description or verified summary is available.

AI Usage: Unknown; assessment pending.

0 stars

Status and source references
  • Below the main-section minimum: 0 stars; 100 required.
  • Assessment pending; source references alone do not establish an app or integration.

Source references: docs/runtime-notes.md

VibeCodyH/pokebench

PokéBench: local and frontier models play Pokémon Red as a benchmark, with a leaderboard and live stream

AI Usage: Unknown; assessment pending.

14 stars

Status and source references
  • Below the main-section minimum: 14 stars; 100 required.
  • Assessment pending; source references alone do not establish an app or integration.

Source references: chatgpt_auth.py

aditnotfound/veil

Veil — invisible desktop AI overlay with BYO keys, local RAG, and Ask/Listen modes (GPL fork of Pluely 0.1.9)

AI Usage: Unknown; assessment pending.

1 stars

Status and source references
  • Below the main-section minimum: 1 stars; 100 required.
  • Assessment pending; source references alone do not establish an app or integration.

Source references: src-tauri/src/chatgpt_plan.rs

aiproxer/aiproxer

AIProxer is a universal LLM proxy that adds an official ChatGPT Plus/Pro plan backend using SIWC and the public OpenAI Responses API.

AI Usage: ChatGPT performs plan-based inference through AIProxer's `openai-chatgpt-plan` backend: users authorize a SIWC profile, the connector obtains a bearer token, and shared Responses transport sends requests to `/v1/responses`. The implementation includes account-specific model discovery and profile-scoped plan inference. [Citations support the connector flow and public endpoints.]

19 stars

Status and source references
  • Below the main-section minimum: 19 stars; 100 required.
  • integration: unresolved. Static authorization, routing, transport, refresh, and inference tests exist, but the claim says the integration is confirmed as a complete usable integration. The mandatory real-harness and sign-off work remains open, and the recorded live model-discovery result parsed zero models, leaving complete current usability unestablished.
  • availability: contradicted. The literal source_available label requires a runnable ChatGPT-plan integration at the target. Although source code and direct live inference exist, the live model-discovery path was recorded as parsing zero models, and the broader acceptance gate remains incomplete. No release artifacts are available either.

Source references: .kiro/specs/openai-chatgpt-plan-connector-migration-spec/openai-chatgpt-plan-connector-migration/design.md, .kiro/specs/openai-chatgpt-plan-connector-migration-spec/openai-chatgpt-plan-connector-migration/requirements.md, .kiro/specs/openai-chatgpt-plan-connector-migration-spec/openai-chatgpt-plan-connector-migration/research.md, .kiro/specs/openai-chatgpt-plan-connector-migration-spec/openai-chatgpt-plan-connector-migration/tasks.md, docs/user_guide/backends/openai-chatgpt-plan.md

alibaba/open-code-review

Secure, fast, efficient, battle-tested at Alibaba's scale. Hybrid architecture code review tool: deterministic pipelines + LLM Agent, precise line-level comments, built-in multi-language ruleset (NPE, thread-safety, XSS, SQL injection), OpenAI & Anthropic compatible.

AI Usage: Unknown; assessment pending.

43352 stars

Status and source references
  • Assessment pending; source references alone do not establish an app or integration.

Source references: Unknown; no source file or discussion collected yet.

anomalyco/opencode

The open source coding agent.

AI Usage: Unknown; assessment pending.

211504 stars

Status and source references
  • Assessment pending; source references alone do not establish an app or integration.

Source references: Unknown; no source file or discussion collected yet.

anshuman-pandey/open-granola

Open-source meeting notebook. Local Whisper microphone transcription, your choice of local or connected AI models, and recoverable summaries in a local library. Tauri + React + Rust. Apache-2.0. Development preview.

AI Usage: Unknown; assessment pending.

10 stars

Status and source references
  • Below the main-section minimum: 10 stars; 100 required.
  • Assessment pending; source references alone do not establish an app or integration.

Source references: src-tauri/src/auth.rs

arturict/tagvico-ai

Self-hosted document automation, automatic tagging and approval-gated companion for Paperless-ngx.

AI Usage: Unknown; assessment pending.

35 stars

Status and source references
  • Below the main-section minimum: 35 stars; 100 required.
  • Assessment pending; source references alone do not establish an app or integration.

Source references: services/chatgptPlanAuthService.ts, tests/chatgpt-plan.test.js

attrition-tech/arkex

A lightweight terminal coding agent for your own models. Connect cloud APIs or local servers to read, edit, and run code with configurable approvals.

AI Usage: Unknown; assessment pending.

0 stars

Status and source references
  • Below the main-section minimum: 0 stars; 100 required.
  • Assessment pending; source references alone do not establish an app or integration.

Source references: internal/chatgpt/chatgpt.go, internal/chatgpt/chatgpt_test.go, internal/chatgpt/login_test.go, internal/provider/chatgpt_test.go, internal/tui/subscription_test.go

aydinmrnv/relay

Local multi-agent software-engineering orchestrator: coordinates Claude Code and Codex to plan, review, implement and critique each other's work on a GitHub issue in an isolated git worktree.

AI Usage: Unknown; assessment pending.

0 stars

Status and source references
  • Below the main-section minimum: 0 stars; 100 required.
  • Assessment pending; source references alone do not establish an app or integration.

Source references: src/auth/chatgpt.ts, test/chatgpt.test.ts

bastani-inc/atomic

The verifiable coding agent runtime. Define your coding agent's process in natural language with stages, checks, and approval gates instead of hoping it follows your instructions. Primitives for verifiable software factories.

AI Usage: Unknown; assessment pending.

843 stars

Status and source references
  • Assessment pending; source references alone do not establish an app or integration.

Source references: packages/ai/src/auth/oauth/openai-chatgpt.ts, packages/ai/test/openai-chatgpt-oauth.test.ts, packages/coding-agent/scripts/assert-pi-runtime-assets.ts

calmmage/botspot

A collection of cross-integrated utils and components for Telegram bots.

AI Usage: Unknown; assessment pending.

1 stars

Status and source references
  • Below the main-section minimum: 1 stars; 100 required.
  • Assessment pending; source references alone do not establish an app or integration.

Source references: Unknown; no source file or discussion collected yet.

can1357/oh-my-pi

⌥ Coding agent with the IDE wired in. Built by Stencil Labs.

AI Usage: Unknown; assessment pending.

34117 stars

Status and source references
  • Assessment pending; source references alone do not establish an app or integration.

Source references: Unknown; no source file or discussion collected yet.

ccch1mneyyy/dsh-TUI

DSH's officially top-recommended TUI plugin — high performance, low overhead, cute pixel whale, smooth mouse interaction. One-command install via npm. / DSH 官方首推的 TUI 插件，高性能低占用，可爱像素鲸鱼，流畅鼠标交互，npm 一键安装

AI Usage: Unknown; assessment pending.

3947 stars

Status and source references
  • Assessment pending; source references alone do not establish an app or integration.

Source references: scripts/verify-oauth.mjs

cgasgarth/clodex

Clodex is a local daemon and CLI that routes OpenAI Codex models into Claude Code using ChatGPT-plan OAuth or API credentials.

AI Usage: ChatGPT-plan OAuth supplies authorization for OpenAI/Codex inference, while Clodex translates Claude Code requests into AI SDK/OpenAI requests and returns generated text or tool calls. This is a direct ChatGPT-plan integration rather than a mere provider mention.

1 stars

Status and source references
  • Below the main-section minimum: 1 stars; 100 required.
  • availability: unresolved. v3.1.0 is clearly released and packaged, but the supplied target-version evidence does not include the complete current OAuth authorization implementation needed to independently verify the entire usable ChatGPT-plan authorization-to-inference chain at that release.
  • chain:authorization: unresolved. The proposed authorization evidence is from commit 7377df1 rather than the recommended v3.1.0 target commit bb8591. The v3.1.0 README documents the sign-in command and plan provider, but the supplied target-version source does not establish issuer, nonce, grant-scope validation, callback handling, and token persistence together.

Source references: src/oauth/openai.ts, tests/daemon-account-command.test.ts, tests/daemon-account-service.test.ts, tests/env-oauth-refresh.test.ts, tests/oauth-credential-store.test.ts

chetto1983/Aura

Local-first, provider-neutral AI agent platform in Go — one binary, graph-backed memory (Postgres + ArcadeDb), self-authored skills, MCP tools, and multi-channel access (CLI / Telegram / web).

AI Usage: Unknown; assessment pending.

2 stars

Status and source references
  • Below the main-section minimum: 2 stars; 100 required.
  • Assessment pending; source references alone do not establish an app or integration.

Source references: internal/chatgptplan/service.go

code-yeongyu/senpi

pi had nothing (nothing), so I made something (something) — sorry mariozechner-senpai, I went ahead and lovingly soiled your pure pi for you. opinionated fork of badlogic/pi-mono with extension-first additions. ganbare ganbare senpi 頑張れ頑張れ先輩

AI Usage: Unknown; assessment pending.

461 stars

Status and source references
  • Assessment pending; source references alone do not establish an app or integration.

Source references: packages/ai/src/auth/oauth/openai-chatgpt.ts, packages/ai/test/openai-chatgpt-oauth.test.ts

coji/asist

A realtime assistant for Mac and Windows. Talk to it and it answers aloud, shows your schedule and mail as cards, and hands longer work to an agent.

AI Usage: Unknown; assessment pending.

0 stars

Status and source references
  • Below the main-section minimum: 0 stars; 100 required.
  • Assessment pending; source references alone do not establish an app or integration.

Source references: Unknown; no source file or discussion collected yet.

dagucloud/dagu

Self-hostable workflow orchestrator for teams whose main work isn't orchestration. Declarative YAML over your scripts, SSH commands, containers, etc; keep workflows separate from business logic. One binary, no database, runs on limited H/W resources. Alternative to Airflow / Cron / Job Scheduler.

AI Usage: Unknown; assessment pending.

4243 stars

Status and source references
  • Assessment pending; source references alone do not establish an app or integration.

Source references: Unknown; no source file or discussion collected yet.

detroitpro/t3code

Unknown; no repository description or verified summary is available.

AI Usage: Unknown; assessment pending.

0 stars

Status and source references
  • Below the main-section minimum: 0 stars; 100 required.
  • Assessment pending; source references alone do not establish an app or integration.

Source references: apps/server/src/provider/CodexChatGptAuth.test.ts, apps/server/src/provider/CodexChatGptAuth.ts, apps/server/src/provider/CodexManagedRuntime.test.ts, apps/server/src/provider/Drivers/CodexDriver.test.ts, apps/server/src/provider/Drivers/CodexManagedProvider.ts

diegopetrucci/the-last-harness

The last harness you'll ever need.

AI Usage: Unknown; assessment pending.

36 stars

Status and source references
  • Below the main-section minimum: 36 stars; 100 required.
  • Assessment pending; source references alone do not establish an app or integration.

Source references: Unknown; no source file or discussion collected yet.

domehahn/harnessmesh

Persistent multi-agent collaboration fabric for agentic coding harnesses. Connect Antigravity, OpenAI Codex, Claude Code and GitHub Copilot through MCP, shared channels, threads, events, evidence and durable state—without human copy/paste.

AI Usage: Unknown; assessment pending.

0 stars

Status and source references
  • Below the main-section minimum: 0 stars; 100 required.
  • Assessment pending; source references alone do not establish an app or integration.

Source references: docs/codex-provider.md, internal/provider/backend_subscription_matrix_test.go, internal/provider/chatgpt_siwc.go, internal/provider/chatgpt_siwc_test.go

dotdioscorea/eyeballs

Requota: native iPhone AI account usage, reset tracking, history and widgets.

AI Usage: Unknown; assessment pending.

0 stars

Status and source references
  • Below the main-section minimum: 0 stars; 100 required.
  • Assessment pending; source references alone do not establish an app or integration.

Source references: App/Eyeballs/OpenAIAuth.swift

eaonlabs/eaon-desktop

Eaon Desktop is a cross-platform AI chat and agent app that can use ChatGPT plans, provider API keys, or local models.

AI Usage: After the user approves Sign in with ChatGPT, Eaon obtains plan-authorized credentials and uses them for streamed Responses API chat, including reasoning and tool calls. The app also supports ordinary BYOK and local-model inference, so ChatGPT-plan usage is one provider path among several.

16 stars

Status and source references
  • Below the main-section minimum: 16 stars; 100 required.
  • license: unresolved. The target package metadata declares MIT, but repository metadata declares GPL-3.0 and the supplied beta license file is GPL-3.0. The evidence therefore conflicts on the applicable license for the proposed target.
  • usefulness: contradicted. The app's capabilities and setup are well evidenced, but the rendered usefulness claim assigns full completeness while the applicable license is materially unresolved due to conflicting MIT and GPL evidence. Thus the complete usefulness field is overstated.

Source references: .eaonbrain/provider-oauth-landscape-sept-2026-which-sign-ins-are-allowed.md, src/main/providers/oauth/siwc.ts, test/provider-signin.test.ts

earendil-works/pi

Pi is an open-source interactive coding-agent CLI and agent harness. Its OpenAI provider supports signing in with a ChatGPT subscription and converting the resulting OAuth access token for OpenAI API use, with refresh-token handling and PKCE-based browser authorization. The integration is released in v1.0.0, but some accounts have reportedly received OpenAI invalid_client errors.

AI Usage: Planned or incomplete integration: Integration is probable rather than confirmed because the source clearly implements subscription OAuth and converts the access token into an API-key-shaped credential, but the supplied evidence does not expose the complete SDK wiring that proves this credential reaches the actual inference request path.

111778 stars

Status and source references
  • chain:inference: unresolved. The evidence shows the OAuth credential being converted to an apiKey-shaped value, but does not show that value reaching a runnable OpenAI inference request path; the proposed claim itself acknowledges this missing wiring.
  • Missing required evidence: End-to-end evidence that the OAuth-derived apiKey is passed through the OpenAI provider into a real inference request.
  • Missing required evidence: Direct v1.0.0 provider setup documentation in the supplied evidence.
  • Missing required evidence: Evidence that OpenAI has enabled the OAuth client flow for all eligible ChatGPT subscription accounts.
  • Confidence: medium. The repository contains a subscription OAuth implementation, tests, and a released v1.0.0 distribution, but the supplied evidence does not show the complete SDK/provider request path from the OAuth credential into an actual inference request. A reported invalid_client failure also indicates availability may vary by account.

Source references: packages/ai/src/auth/oauth/openai-chatgpt.ts, packages/ai/test/openai-chatgpt-oauth.test.ts

edlundin/kiln

Unknown; no repository description or verified summary is available.

AI Usage: Unknown; assessment pending.

0 stars

Status and source references
  • Below the main-section minimum: 0 stars; 100 required.
  • Assessment pending; source references alone do not establish an app or integration.

Source references: docs/learning/edl-342.md

epicycloids/multivac-client

Contributor client, project protocol, and independent research endpoint starter for Multivac.

AI Usage: Unknown; assessment pending.

0 stars

Status and source references
  • Below the main-section minimum: 0 stars; 100 required.
  • Assessment pending; source references alone do not establish an app or integration.

Source references: src/research_market/chatgpt_plan.py

eunsoogi/orot

Unknown; no repository description or verified summary is available.

AI Usage: Unknown; assessment pending.

1 stars

Status and source references
  • Below the main-section minimum: 1 stars; 100 required.
  • Assessment pending; source references alone do not establish an app or integration.

Source references: Unknown; no source file or discussion collected yet.

everruns/yolop

Yolop is a MIT-licensed terminal coding agent with a documented, source-available ChatGPT-plan integration. Users can opt into the open-source Sign in with ChatGPT route, which performs browser OAuth with PKCE, requires the ChatGPT plan-usage scope, and sends turns through the public Responses API. The integration is available from the pinned source commit but is not established as included in the v0.18.4 release.

AI Usage: For an open-source login, the driver uses the public Responses API, shapes the request for that API, sends the bearer token, and converts unsuccessful responses through the ChatGPT-plan error path. This connects the consented plan login to actual inference rather than merely storing credentials.

16 stars

Status and source references
  • Below the main-section minimum: 16 stars; 100 required.
  • chain:distribution: unresolved. The recommendation explicitly makes no distribution claim for the 798e849 source target, and the supplied release evidence is for a different commit; source availability does not establish packaging in a released artifact.
  • Missing required evidence: No supplied evidence proves that the open-source ChatGPT integration was packaged into a released binary or Homebrew artifact at commit 798e8498c7a82839ee45d55c00e4aff1520b817d.
  • Missing required evidence: No live-account evidence establishes that the preview route works end to end against production ChatGPT accounts.

Source references: knowledge/specs/chatgpt-sign-in.md, src/auth/siwc.rs, src/config/mod.rs, src/drivers/chatgpt_plan.rs

ferraridavide/decisionwise

Decide with confidence.

AI Usage: Unknown; assessment pending.

1 stars

Status and source references
  • Below the main-section minimum: 1 stars; 100 required.
  • Assessment pending; source references alone do not establish an app or integration.

Source references: server/chatgpt.ts

gannonh/kata-code

Cloud, desktop and mobile coding agent orchestration

AI Usage: Unknown; assessment pending.

1 stars

Status and source references
  • Below the main-section minimum: 1 stars; 100 required.
  • Assessment pending; source references alone do not establish an app or integration.

Source references: apps/server/src/provider/CodexChatGptAuth.test.ts, apps/server/src/provider/CodexChatGptAuth.ts, apps/server/src/provider/CodexManagedRuntime.test.ts, apps/server/src/provider/Drivers/CodexDriver.test.ts, apps/server/src/provider/Drivers/CodexManagedProvider.ts

gethamster/horde

Horde provides a released direct Sign in with ChatGPT integration for managed accounts. It supports OAuth authorization with PKCE and loopback callbacks, validates signed identity tokens, stores credentials in protected managed-account storage, refreshes tokens, exposes ChatGPT models, and lets native workers execute tasks using the authorized ChatGPT plan.

AI Usage: The authorized account is wired into a ChatGPT Provider with auth_mode login and account binding; the native executor then performs the ChatGPT request flow and completes the worker task, establishing the actual inference call path rather than merely proving configuration ancestry.

10 stars

Status and source references
  • Below the main-section minimum: 10 stars; 100 required.
  • Missing required evidence: No general end-user setup documentation or configuration-file example was supplied beyond the implementation and tests.
  • Missing required evidence: The evidence demonstrates macOS and Linux release assets but does not establish Windows support.

Source references: src/chatgpt_auth.rs, src/chatgpt_auth/integration_tests.rs, src/chatgpt_auth/session.rs, src/chatgpt_auth/storage.rs, tests/chatgpt_boundaries.rs

ggfincke/456code

Personal fork of t3code (https://github.com/pingdotgg/t3code) — one control plane for agentic development

AI Usage: Unknown; assessment pending.

1 stars

Status and source references
  • Below the main-section minimum: 1 stars; 100 required.
  • Assessment pending; source references alone do not establish an app or integration.

Source references: apps/server/src/provider/CodexChatGptAuth.test.ts, apps/server/src/provider/CodexChatGptAuth.ts, apps/server/src/provider/CodexManagedRuntime.test.ts, apps/server/src/provider/Drivers/CodexDriver.test.ts, apps/server/src/provider/Drivers/CodexManagedProvider.ts

gitroomhq/postiz-app

📨 The ultimate agentic social media scheduling tool 🤖

AI Usage: Unknown; assessment pending.

36637 stars

Status and source references
  • Assessment pending; source references alone do not establish an app or integration.

Source references: Unknown; no source file or discussion collected yet.

gudyfut/GudyBrain

GudyBrain is a local personal assistant with Markdown memory, text and voice chat, and Discord call analysis.

AI Usage: ChatGPT generates conversational responses through the Responses API after the user connects and authorizes a ChatGPT plan; Jev selects relevant memory and the application supplies the resulting context. The same integration supports structured model output for memory navigation and other orchestration tasks.

0 stars

Status and source references
  • Below the main-section minimum: 0 stars; 100 required.
  • platforms: unresolved. Windows and Linux/macOS instructions are supplied for the Discord companion, but the evidence does not establish that the main web application itself is supported on all three platforms.
  • chain:distribution: unresolved. There is no release artifact or packaging evidence for the web application. The source recommendation does not require such distribution support, but the distribution chain itself is not established.

Source references: src/core/chatgpt-auth.ts, src/dev/validate-responses.ts

hafiezul/rove

An open-source workspace for coding agents. Bring your own subscriptions, work across web, desktop, and mobile, and connect remotely.

AI Usage: Unknown; assessment pending.

0 stars

Status and source references
  • Below the main-section minimum: 0 stars; 100 required.
  • Assessment pending; source references alone do not establish an app or integration.

Source references: apps/server/src/provider/CodexChatGptAuth.test.ts, apps/server/src/provider/CodexChatGptAuth.ts, apps/server/src/provider/CodexManagedRuntime.test.ts, apps/server/src/provider/Drivers/CodexDriver.test.ts, apps/server/src/provider/Drivers/CodexManagedProvider.ts

hwkwon1114/pi-setup

Pi Setup is a portable configuration and extension bundle for the Pi coding agent, including multi-account ChatGPT OAuth support and research tools.

AI Usage: Pi uses ChatGPT OAuth accounts as selectable OpenAI providers for coding-agent model requests, with multi-turn streaming, account switching, and 429 failover. It also displays ChatGPT/Codex usage information where supported, while direct subscription allowance remains unknown unless a quota block is observed.

0 stars

Status and source references
  • Below the main-section minimum: 0 stars; 100 required.

Source references: README.md, extensions/multi-openai/README.md, extensions/test-usage-limits.mjs, extensions/usage-limits.ts

iivankin/oauthproxy

OAuth Proxy is a Bun server that lets clients use ChatGPT plan sharing through OpenAI’s Responses API over HTTP/SSE and WebSocket routes.

AI Usage: ChatGPT performs Responses API inference for client-supplied models and conversation inputs, with the proxy handling OAuth plan consent, token refresh, model discovery, streaming, and WebSocket forwarding. The integration is specifically authorized for ChatGPT plan usage through the `chatgpt.tokens.use.direct` scope.

0 stars

Status and source references
  • Below the main-section minimum: 0 stars; 100 required.

Source references: src/chatgpt/accounts.ts, src/chatgpt/oauth.ts, src/chatgpt/types.ts, test/chatgpt.test.ts

inclunet/assistente

Assistente pessoal com arquitetura de agentes inteligentes - Go/Wails/Svelte

AI Usage: Unknown; assessment pending.

2 stars

Status and source references
  • Below the main-section minimum: 2 stars; 100 required.
  • Assessment pending; source references alone do not establish an app or integration.

Source references: internal/oauthintegrations/chatgpt.go

isaachorowitz/TapLyne-App

An open-source Mac app that gives AI agents a real iPhone through local USB capture and Bluetooth input.

AI Usage: Unknown; assessment pending.

0 stars

Status and source references
  • Below the main-section minimum: 0 stars; 100 required.
  • Assessment pending; source references alone do not establish an app or integration.

Source references: Taplyne/Agent/ChatGPTAuth.swift

jaja360/kat-irc

kat-irc is a configurable Go IRC assistant that uses OpenAI models to answer addressed channel messages and optionally react with emoji.

AI Usage: The bot sends addressed IRC messages, persona, memory, and bounded channel history to the OpenAI Responses API for streamed replies, and can use a reaction tool for IRCv3 emoji reactions. ChatGPT OAuth connects those inference requests to an eligible ChatGPT plan's shared usage limits rather than API billing.

0 stars

Status and source references
  • Below the main-section minimum: 0 stars; 100 required.

Source references: README.md, oauth.go

jal-co/pi-chatgpt-pool

A pi extension that pools multiple ChatGPT accounts and fails over between them when an account is limited.

AI Usage: ChatGPT performs the selected model inference through signed-in ChatGPT accounts; the extension routes each request to an available account and retries on another account after a limit. This is connected to pi's ChatGPT sign-in and pooled model flow, although the supplied excerpts do not show the underlying SDK call wiring.

0 stars

Status and source references
  • Below the main-section minimum: 0 stars; 100 required.
  • integration: unresolved. The README and package metadata describe a pi ChatGPT integration and declare an extension entry point, but the supplied evidence does not show the runnable authorization-to-inference wiring needed to independently confirm the complete integration.
  • availability: unresolved. A non-prerelease v0.2.3 release, package metadata, and an install command are evidenced, but the complete usable ChatGPT plan integration at the target is not independently established because the authorization-to-inference wiring is absent from the supplied excerpts.
  • usefulness: unresolved. Account pooling, routing, and failover are documented and plausibly useful, but the rendered perfect scores assert continued working inference and complete capability; the supplied evidence does not independently establish the complete runnable authorization-to-inference chain.
  • chain:authorization: unresolved. The evidence establishes account addition and pi's Sign in with ChatGPT flow, but does not show how the resulting authorization is wired into the inference request path.
  • chain:inference: unresolved. The README documents model selection, request routing, and failover, but the supplied source excerpts omit the extension implementation or direct inference SDK call needed to verify usable inference wiring.
  • Missing required evidence: Direct source evidence connecting pi's authorization SDK output to the ChatGPT inference request path is not included in the supplied excerpts.
  • Confidence: medium. The pinned README documents ChatGPT sign-in, model selection, account pooling, and failover, while package metadata identifies a pi extension and the supplied target has a non-prerelease release. Confidence is reduced because the supplied source excerpts do not expose the extension's direct SDK inference wiring.

Source references: src/usage.ts, test/usage.test.ts

jckail/point_bot

Unknown; no repository description or verified summary is available.

AI Usage: Unknown; assessment pending.

1 stars

Status and source references
  • Below the main-section minimum: 1 stars; 100 required.
  • Assessment pending; source references alone do not establish an app or integration.

Source references: docs/chatgpt-standalone-plan.md

justrach/harness

Harness desktop app source prepared for public release

AI Usage: Unknown; assessment pending.

3 stars

Status and source references
  • Below the main-section minimum: 3 stars; 100 required.
  • Assessment pending; source references alone do not establish an app or integration.

Source references: Unknown; no source file or discussion collected yet.

jvm/pi-mono

Monorepo for Pi-related projects

AI Usage: Unknown; assessment pending.

28 stars

Status and source references
  • Below the main-section minimum: 28 stars; 100 required.
  • Assessment pending; source references alone do not establish an app or integration.

Source references: packages/pi-fast/tests/native-openai-smoke.mjs

k3dom/pi-plugins

Pi-plugins is a collection of installable pi extensions for web search, page fetching, delegation, and workflow utilities.

AI Usage: Pi can use a ChatGPT subscription by obtaining an OAuth access token through Sign in with ChatGPT and passing it as the OpenAI Responses API credential. The host then sends model requests and tool-enabled context through the OpenAI SDK; the exact websearch-to-provider registration is not shown in the supplied evidence.

13 stars

Status and source references
  • Below the main-section minimum: 13 stars; 100 required.
  • integration: contradicted. Host-level ChatGPT OAuth and OpenAI Responses wiring is present, but the proposed rendered integration is marked confirmed for the websearch recommendation while the evidence does not show websearch registering or enabling that provider. The supplied missing-link statement itself identifies this gap.
  • availability: contradicted. A released websearch package is evidenced, but the rendered availability claim treats the assessed ChatGPT/websearch integration as released. The complete usable package-level integration chain is not shown, so the release of the base package cannot establish availability of that integration.
  • setup: unresolved. Pi version, plugin loading, and the host OAuth browser/callback procedure are documented, but no evidence gives a complete setup path showing the released websearch package itself enabling the ChatGPT provider.
  • chain:inference: unresolved. The host source clearly constructs an OpenAI Responses client and sends model requests with transcript and tool context, but the evidence does not establish that the released websearch package wires this authorization into the provider path.
  • Missing required evidence: Explicit source showing the websearch package registering or enabling the OpenAI ChatGPT provider at the assessed release target.
  • Confidence: medium. The repository documents installable pi plugins and the supplied source includes a complete Sign in with ChatGPT OAuth flow plus OpenAI Responses API wiring, but the evidence does not explicitly show the websearch package itself registering the provider.

Source references: .agents/repos/pi/packages/ai/src/auth/oauth/openai-chatgpt.ts, .agents/repos/pi/packages/ai/test/openai-chatgpt-oauth.test.ts

kkestell/ox

Unknown; no repository description or verified summary is available.

AI Usage: Unknown; assessment pending.

0 stars

Status and source references
  • Below the main-section minimum: 0 stars; 100 required.
  • Assessment pending; source references alone do not establish an app or integration.

Source references: crates/ox-server/src/openai_auth.rs

klovr-co/hover-tag

Your personal assistant in Slack, powered by your Codex or Claude Code subscription. One for every teammate. Runs on a computer you control.

AI Usage: Unknown; assessment pending.

7 stars

Status and source references
  • Below the main-section minimum: 7 stars; 100 required.
  • Assessment pending; source references alone do not establish an app or integration.

Source references: scripts/tag_chatgpt.py

kokotatan/katazuku-shukatsu

katazuku-shukatsu is a local-first Japanese job-hunting autopilot that uses AI to organize workflows while leaving consequential decisions to the user.

AI Usage: ChatGPT can perform tool-free job-hunting tasks such as extracting selection updates from email, while katazuku authenticates the user through Sign in with ChatGPT and sends approved prompts to the OpenAI Responses API. Tool-dependent workflows remain assigned to providers with tool access.

2 stars

Status and source references
  • Below the main-section minimum: 2 stars; 100 required.

Source references: docs/AI-PROVIDERS.md, src/providers/chatgpt-siwc.ts, tests/check-providers.ts

kongdd/ai-quota

AI coding Plan额度查询: OpenAI, Claude, Grok, Opencode-go, Minimax, DeepSeek, Kimi; earthengine

AI Usage: Unknown; assessment pending.

0 stars

Status and source references
  • Below the main-section minimum: 0 stars; 100 required.
  • Assessment pending; source references alone do not establish an app or integration.

Source references: src/provider/openai.ts, test/openai-apps.test.mjs, test/openai.test.mjs

ksgat/zither

Unknown; no repository description or verified summary is available.

AI Usage: Unknown; assessment pending.

0 stars

Status and source references
  • Below the main-section minimum: 0 stars; 100 required.
  • Assessment pending; source references alone do not establish an app or integration.

Source references: electron/chatgpt.ts

kzinmr/ai-topics-pi

Unknown; no repository description or verified summary is available.

AI Usage: Unknown; assessment pending.

0 stars

Status and source references
  • Below the main-section minimum: 0 stars; 100 required.
  • Assessment pending; source references alone do not establish an app or integration.

Source references: tests/sdk.test.mjs

ldbumble/taskuary

Automate your job: local-first AI task hub. Email, Teams, Slack & reports -> one timeline -> AI triage -> your coding agents (Claude Code, Codex, Gemini) do the work, you approve.

AI Usage: Unknown; assessment pending.

135 stars

Status and source references
  • Assessment pending; source references alone do not establish an app or integration.

Source references: taskuary/chatgptauth.py, tests/test_chatgptauth.py

ldclabs/anda-bot

🐼 Born of panda. Awakened as Anda. A copilot for your computer, powered by the Anda Brain.

AI Usage: Unknown; independent verification pending.

26 stars

Status and source references
  • Below the main-section minimum: 26 stars; 100 required.
  • Independent verification pending; recommendation unresolved.

Source references: anda_bot/src/chatgpt/mod.rs

leandromesq/lelezonio-pi-kit

Clone-based Pi coding agent setup with local extensions, subagents, workflows, git tools, and a custom TUI

AI Usage: Unknown; assessment pending.

0 stars

Status and source references
  • Below the main-section minimum: 0 stars; 100 required.
  • Assessment pending; source references alone do not establish an app or integration.

Source references: extensions/codex-accounts/account-store.test.ts

leynier/alera

A native, performance-first ADE built with Flutter, Rust, and Ghostty. Run Claude Code, Codex, Amp, and other CLI coding agents in parallel. No Electron.

AI Usage: Unknown; assessment pending.

36 stars

Status and source references
  • Below the main-section minimum: 36 stars; 100 required.
  • Assessment pending; source references alone do not establish an app or integration.

Source references: rust/alera-cli/src/terminal_host/server/chatgpt_credentials.rs, rust/alera-cli/src/terminal_host/server/chatgpt_oauth.rs, rust/alera-cli/src/terminal_host/server/chatgpt_session_tests.rs, rust/alera-cli/src/terminal_host/server/chatgpt_tests.rs

leynier/exeora

Secure execution for AI agents, on any machine. Connect any MCP client to the development environment on any machine you can run a command on (even your own laptop) - no open ports, no tunnels, and your source code never leaves the machine.

AI Usage: Unknown; assessment pending.

7 stars

Status and source references
  • Below the main-section minimum: 7 stars; 100 required.
  • Assessment pending; source references alone do not establish an app or integration.

Source references: apps/gateway/src/ai/chatgpt.workers.test.ts, apps/web/landing/src/pages/docs/ai-assist.astro, crates/exeora-cli/readme.md, crates/exeora-cli/src/chatgpt/inference.rs, crates/exeora-cli/src/chatgpt/mod.rs

llms-txt-archive/openai-platform

Versioned mirror of OpenAI developer docs from llms.txt

AI Usage: Unknown; assessment pending.

1 stars

Status and source references
  • Below the main-section minimum: 1 stars; 100 required.
  • Assessment pending; source references alone do not establish an app or integration.

Source references: cookbook/articles/sign-in-with-chatgpt.md, siwc/token-sharing-open-source/errors-and-recovery.md, siwc/token-sharing-open-source/sign-in.md, siwc/token-sharing-open-source/token-reference.md

longyijdos/kana

Lightweight, untracked, and self-evolving terminal AI companion. Zero telemetry, durable memory, bidirectional synergy, and native TUI.

AI Usage: Unknown; assessment pending.

8 stars

Status and source references
  • Below the main-section minimum: 8 stars; 100 required.
  • Assessment pending; source references alone do not establish an app or integration.

Source references: docs/openai-codex-provider.md, docs/openai-codex-provider.zh-CN.md, src/kana/auth/openai-codex.ts, tests/kana/auth/openai-codex.test.ts

looprig/carbon

Unknown; no repository description or verified summary is available.

AI Usage: Unknown; assessment pending.

1 stars

Status and source references
  • Below the main-section minimum: 1 stars; 100 required.
  • Assessment pending; source references alone do not establish an app or integration.

Source references: internal/app/subscription_test.go

luckydye/vektor

Unknown; no repository description or verified summary is available.

AI Usage: Unknown; assessment pending.

4 stars

Status and source references
  • Below the main-section minimum: 4 stars; 100 required.
  • Assessment pending; source references alone do not establish an app or integration.

Source references: server/docs/extensions.md

lxorb/nibeditor

nibeditor is an open-source Markdown editor and note-taking app with sync, MCP, and AI-assisted note features.

AI Usage: Unknown; AI usage is not independently verified.

3 stars

Status and source references
  • Below the main-section minimum: 3 stars; 100 required.
  • ai_usage: unresolved. The evidence supports ChatGPT-plan authorization and says the release uses a ChatGPT plan for AI features, but it does not establish that the obtained token is actually wired to the claimed note-question and note-edit inference operations in the desktop application.
  • integration: unresolved. Authorization, token issuance, refresh, and the Responses API resource are implemented, but the supplied evidence does not show a complete call from the app's AI features through chatgpt_token to an actual Responses API inference request.
  • availability: unresolved. A stable v0.11.0 release with desktop assets exists, but availability of the claimed usable ChatGPT-plan integration is not established because the complete inference wiring is missing from the supplied evidence.
  • setup: unresolved. The sign-in setup flow is documented and implemented through browser authorization and a loopback callback, but the evidence does not establish that completing it enables the claimed AI features end to end.
  • usefulness: unresolved. The app's notes, MCP, sync, and AI capabilities are useful and supported, but the specific usefulness claim depends on direct ChatGPT-plan inference being usable for note operations, which the evidence does not fully establish.
  • chain:inference: unresolved. The access token is exposed by a desktop command and refreshed for the Responses API resource, but no supplied source shows the actual AI feature invoking that command and sending an inference request.
  • chain:distribution: unresolved. The exact target has a stable release and platform assets, but packaging inclusion of a complete usable authorization-to-inference integration is not established because the inference wiring is absent from the supplied evidence.

Source references: apps/desktop/src-tauri/src/chatgpt/oauth.rs

m-rui001/MATE

Unknown; no repository description or verified summary is available.

AI Usage: Unknown; assessment pending.

1 stars

Status and source references
  • Below the main-section minimum: 1 stars; 100 required.
  • Assessment pending; source references alone do not establish an app or integration.

Source references: packages/ai/src/auth/oauth/openai-chatgpt.ts

maxprain12/dome

Unknown; no repository description or verified summary is available.

AI Usage: Unknown; assessment pending.

23 stars

Status and source references
  • Below the main-section minimum: 23 stars; 100 required.
  • Assessment pending; source references alone do not establish an app or integration.

Source references: packages/ai/src/auth/oauth/openai-chatgpt.ts

mhismail3/tron

Unknown; no repository description or verified summary is available.

AI Usage: Unknown; assessment pending.

0 stars

Status and source references
  • Below the main-section minimum: 0 stars; 100 required.
  • Assessment pending; source references alone do not establish an app or integration.

Source references: packages/gateway/src/admin/auth-broker.test.ts

misaon/workbox

Local-first, TypeScript-first orchestrator for AI coding agents, with a pixel-art office where every agent is an employee

AI Usage: Unknown; assessment pending.

0 stars

Status and source references
  • Below the main-section minimum: 0 stars; 100 required.
  • Assessment pending; source references alone do not establish an app or integration.

Source references: docs/superpowers/research/2026-09-30-R1-subscriptions-and-agent-auth.md

miuuyy/Clew

Clear path from where you are to what you want to learn. AI agent turns goals and topic dumps into graph with visible dependencies (generative roadmap.sh)

AI Usage: Unknown; assessment pending.

187 stars

Status and source references
  • Assessment pending; source references alone do not establish an app or integration.

Source references: backend/app/agent/chatgpt_auth.py, backend/tests/fake_chatgpt.py, backend/tests/test_chatgpt_auth.py, docs/ARCHITECTURE.md, docs/adr/0006-sign-in-with-chatgpt-agent-runtime.md

miziusLabs/Dray

Unknown; no repository description or verified summary is available.

AI Usage: Unknown; assessment pending.

1 stars

Status and source references
  • Below the main-section minimum: 1 stars; 100 required.
  • Assessment pending; source references alone do not establish an app or integration.

Source references: apps/desktop/src-tauri/src/account.rs

neavo/LinguaGacha

使用 AI 能力一键翻译 小说、游戏、字幕 等文本内容的次世代文本翻译器

AI Usage: Unknown; ChatGPT usage is not established.

2527 stars

Status and source references
  • summary: unresolved. The application purpose, v0.125.0 release, ChatGPT login, Codex Plan quota, and GPT Fast/UltraFast modes are supported, but the rendered claim that the authenticated credential is routed into the inference API is not fully established because the evidence does not show the ChatGPT request adapter being registered in the runtime path.
  • availability: unresolved. A non-prerelease v0.125.0 distribution exists and announces ChatGPT Login, but the literal claim that the released app provides the complete usable ChatGPT authorization-to-inference feature depends on runtime wiring that is not fully shown by the supplied evidence.
  • categories: unresolved. Translation, novel, game, subtitle, and AI-assisted workflows are evidenced, but the complete category list includes the broader label 'AI writing' and the precise 'local desktop app' classification without direct supporting evidence. The rendered multi-category claim is therefore only partially established.
  • chain:inference: unresolved. The evidence shows OAuth credentials converted to an API key and separately shows a ChatGPT request adapter that shapes requests and observes failures, but it does not show the adapter being wired into the authenticated model/provider execution path. The usable authorization-to-inference chain is therefore incomplete.
  • Open-source licensing is not established.

Source references: src/backend/auth/chatgpt-auth-service.test.ts, src/backend/auth/chatgpt-oauth.test.ts, src/backend/auth/chatgpt-oauth.ts

neuroforge-io/Sinter

Free, local-first community workbench for funding research, meeting records and handovers, with source-linked work and optional AI assistance.

AI Usage: Unknown; assessment pending.

0 stars

Status and source references
  • Below the main-section minimum: 0 stars; 100 required.
  • Assessment pending; source references alone do not establish an app or integration.

Source references: docs/PROVIDER_READINESS_2026-09-30.md, src/sinter/accounts.py

nimiplatform/nimi

AI-native open world platform — runtime, SDK, desktop & mods

AI Usage: Unknown; assessment pending.

5 stars

Status and source references
  • Below the main-section minimum: 5 stars; 100 required.
  • Assessment pending; source references alone do not establish an app or integration.

Source references: .nimi/spec/runtime/ai-provider.authority.yaml, apps/desktop/test/electron-connector-auth-acquisition-host.test.ts, config/sdks-connector-auth-acquisition-profiles.yaml, runtime/internal/services/ai/chatgpt_plan_app_text_test.go, runtime/internal/services/connector/chatgpt_plan_credential.go

nourdean22/MAINnicks-tire-autoNEW

Professional website for Nick's Tire & Auto, a high-volume independent tire and auto repair shop at 17625 Euclid Ave, Cleveland, OH 44112. Features service pages, customer reviews, about page, contact form, and blog for local SEO. Designed to replace the current single-page Google Sites placeholder.

AI Usage: Unknown; assessment pending.

1 stars

Status and source references
  • Below the main-section minimum: 1 stars; 100 required.
  • Assessment pending; source references alone do not establish an app or integration.

Source references: apps/statenour/local-agent/chatgpt-plan-bridge.mjs

ogulcancelik/pi-extensions

Extensions for pi, the terminal-based coding agent

AI Usage: Unknown; assessment pending.

553 stars

Status and source references
  • Assessment pending; source references alone do not establish an app or integration.

Source references: Unknown; no source file or discussion collected yet.

openai/sign-in-with-chatgpt-devkit

An OpenAI developer kit for adding Sign in with ChatGPT and ChatGPT-plan-powered AI features to local apps.

AI Usage: With user permission, the SDK signs users in, obtains credentials including the direct token-sharing scope, and sends streaming Responses API requests using the user's ChatGPT plan. Applications can use this for text transformation and other AI features without asking users for a separate API key.

17 stars

Status and source references
  • Below the main-section minimum: 17 stars; 100 required.
  • Open-source licensing is not established.

Source references: examples/paste-perfect/test/credential-storage-runtime-worker.mjs, packages/local/src/index.ts, packages/local/src/oauth.ts, packages/local/test/refresh.test.mjs, packages/local/test/storage.test.mjs

openclaw-easy/openclaw-easy-desktop

OpenClaw Desktop App — No Terminal, No Coding. One-click installer for macOS & Windows.

AI Usage: Unknown; assessment pending.

22 stars

Status and source references
  • Below the main-section minimum: 22 stars; 100 required.
  • Assessment pending; source references alone do not establish an app or integration.

Source references: openclaw/extensions/openai/token-sharing-oauth.runtime.test.ts, openclaw/extensions/openai/token-sharing-oauth.runtime.ts, openclaw/extensions/openai/token-sharing.ts

orbi-build/orbi

Open-source (AGPL-3.0), self-hosted autonomous coding agent: label a GitHub Issue, get an independently reviewed, merged PR and a tagged release.

AI Usage: Unknown; assessment pending.

194 stars

Status and source references
  • Assessment pending; source references alone do not establish an app or integration.

Source references: docs/providers.mdx, docs/zh/providers.mdx, tests/test_docs_site.py

pantoniou/fyai

fyai is a C-based Unix coding agent with durable local conversation state, provider-independent model support, tools, branching, and a ChatGPT subscription route using OpenAI's direct token-sharing flow. ChatGPT access is limited to eligible OpenAI Responses requests and requires explicit subscription authorization.

AI Usage: After authorization, fyai resolves ChatGPT mode for eligible OpenAI Responses configurations, sets the Responses endpoint and streaming behavior, and applies the saved bearer token to inference requests. The documented inference path uses stream:true and store:false.

42 stars

Status and source references
  • Below the main-section minimum: 42 stars; 100 required.
  • Missing required evidence: No stable, non-prerelease release containing this integration is evidenced.
  • Missing required evidence: The supplied release metadata is dated 2026-07-08, so the current real-world availability of that prerelease cannot be independently established from the repository evidence alone.

Source references: doc/chatgpt-auth.md, src/fyai_auth.c, tests/cases/auth_config.sh, tests/cases/auth_registration.sh, tests/cases/transport_login_commands.sh

pejmanjohn/chickpea

Self-hosted, model-agnostic AI teammates in Slack. Agents with their own @handle, memory, skills, and per-channel permissions. 35 connector presets (Zendesk, Stripe, GitHub, HubSpot), MCP and API connections, scheduled work, coding sandbox. Manage from Slack, Admin, or MCP. Built on Flue. Cloudflare Workers or Node. MIT.

AI Usage: Unknown; assessment pending.

7 stars

Status and source references
  • Below the main-section minimum: 7 stars; 100 required.
  • Assessment pending; source references alone do not establish an app or integration.

Source references: src/chatgpt-plan/protocol.ts, tests/chatgpt-plan.test.ts

pingdotgg/t3code

Unknown; no repository description or verified summary is available.

AI Usage: Unknown; assessment pending.

24328 stars

Status and source references
  • Assessment pending; source references alone do not establish an app or integration.

Source references: apps/server/src/provider/CodexChatGptAuth.test.ts, apps/server/src/provider/CodexChatGptAuth.ts, apps/server/src/provider/CodexManagedRuntime.test.ts, apps/server/src/provider/Drivers/CodexDriver.test.ts, apps/server/src/provider/Drivers/CodexManagedProvider.ts

porarrirr/yamabikochat

YamabikoChat is a native Android and iOS multi-provider AI chat app with model comparison, automatic conversations, attachments, and ChatGPT plan support.

AI Usage: ChatGPT performs authenticated model discovery and Responses API inference, including streamed replies and namespaced local tool calls. The app obtains ChatGPT plan consent through Pi's SIWC OAuth provider and passes the resulting credential into the Responses adapter.

0 stars

Status and source references
  • Below the main-section minimum: 0 stars; 100 required.

Source references: app/src/main/java/com/porarri/yamabikochat/data/auth/CodexAuthRepository.kt, app/src/test/java/com/porarri/yamabikochat/data/auth/CodexAuthRepositoryTest.kt, ios/PiRuntime/src/chatgpt-plan-plugin.js, ios/PiRuntime/test/chatgpt-plan-bridge.test.js, ios/PiRuntime/test/chatgpt-plan-plugin.test.js

pwrdrvr/PwrGit

Manage hundreds of Git repos and thousands of worktrees with ease. Git your agents and git work done!

AI Usage: Unknown; assessment pending.

3 stars

Status and source references
  • Below the main-section minimum: 3 stars; 100 required.
  • Assessment pending; source references alone do not establish an app or integration.

Source references: Unknown; no source file or discussion collected yet.

racetozero/kiss

KISS is a Rust terminal coding-agent application with a shipped ChatGPT subscription integration. Users can install the released binary, run `kiss login openai`, authenticate in a browser, and use OpenAI models with agent tools, sessions, automation, and MCP support.

AI Usage: The authenticated credential is wired into the exact inference path: the Responses adapter reads the credential, identifies OpenAI Codex, derives the ChatGPT account ID, builds the request, and sends it with bearer authentication and ChatGPT account headers.

18 stars

Status and source references
  • Below the main-section minimum: 18 stars; 100 required.

Source references: crates/kiss-ai/src/auth/openai_codex.rs

rakanssh/hakawati

Hakawati is a desktop AI adventure client for playing interactive tales with configurable AI providers, including ChatGPT subscriptions.

AI Usage: ChatGPT can narrate tales and provide utility-model functions such as quickstart, scenario generation, and story-card generation. Hakawati obtains ChatGPT OAuth credentials, requires the plan-usage scope, and sends streaming Responses API requests using the connected ChatGPT plan.

20 stars

Status and source references
  • Below the main-section minimum: 20 stars; 100 required.

Source references: src-tauri/src/chatgpt/mod.rs, src-tauri/src/chatgpt/vault.rs

rcarmo/go-ai

A mildly sane inference API library for go

AI Usage: Unknown; assessment pending.

7 stars

Status and source references
  • Below the main-section minimum: 7 stars; 100 required.
  • Assessment pending; source references alone do not establish an app or integration.

Source references: oauth/openai_chatgpt.go

rcarmo/piclaw

pi coding agent in a technicolor web trenchcoat

AI Usage: Unknown; assessment pending.

867 stars

Status and source references
  • Assessment pending; source references alone do not establish an app or integration.

Source references: runtime/test/agent-control/fixtures/cli-auth-preload-0991.mjs, runtime/test/agent-control/fixtures/cli-auth-preload-100.mjs, runtime/test/agent-control/fixtures/packaged-cli-auth-0991.ts, runtime/test/agent-control/fixtures/packaged-cli-auth-100.ts, runtime/test/agent-control/fixtures/packaged-openai-login-0991.mjs

rcarmo/rs-ai

Unified LLM API — Rust port of @earendil-works/pi-ai

AI Usage: Unknown; assessment pending.

2 stars

Status and source references
  • Below the main-section minimum: 2 stars; 100 required.
  • Assessment pending; source references alone do not establish an app or integration.

Source references: src/openai_chatgpt_oauth.rs

riffado/riffado

Self-hosted AI transcription interface for Plaud Note devices

AI Usage: Unknown; assessment pending.

392 stars

Status and source references
  • Assessment pending; source references alone do not establish an app or integration.

Source references: Unknown; no source file or discussion collected yet.

rmonvfer/micro

A terminal coding agent with local, inspectable session logs.

AI Usage: Unknown; assessment pending.

0 stars

Status and source references
  • Below the main-section minimum: 0 stars; 100 required.
  • Assessment pending; source references alone do not establish an app or integration.

Source references: crates/micro-auth/src/chatgpt.rs

rubenfb23/research-pi

Open-source research harness on the Pi SDK. Reproducible ML experiments, scientific protocols and evidence-linked papers. CLI: repi.

AI Usage: Unknown; assessment pending.

2 stars

Status and source references
  • Below the main-section minimum: 2 stars; 100 required.
  • Assessment pending; source references alone do not establish an app or integration.

Source references: tests/connections.test.ts

sequelcore/tesota

An open-source, verification-first agent for work that carries its evidence

AI Usage: Unknown; assessment pending.

2 stars

Status and source references
  • Below the main-section minimum: 2 stars; 100 required.
  • Assessment pending; source references alone do not establish an app or integration.

Source references: Unknown; no source file or discussion collected yet.

servoagents/entryplug

Entryplug is an agent embodiment exokernel

AI Usage: Unknown; assessment pending.

0 stars

Status and source references
  • Below the main-section minimum: 0 stars; 100 required.
  • Assessment pending; source references alone do not establish an app or integration.

Source references: src/entryplug_agents/auth.py

sethhen/t3code

Personal fork of pingdotgg/t3code with automatic account pooling & opinionated improvements. Will be kept up-to-date with T3.

AI Usage: Unknown; assessment pending.

2 stars

Status and source references
  • Below the main-section minimum: 2 stars; 100 required.
  • Assessment pending; source references alone do not establish an app or integration.

Source references: apps/server/src/provider/CodexChatGptAuth.test.ts, apps/server/src/provider/CodexChatGptAuth.ts, apps/server/src/provider/CodexManagedRuntime.test.ts, apps/server/src/provider/Drivers/CodexDriver.test.ts

shriism/serenity

Unknown; no repository description or verified summary is available.

AI Usage: Unknown; assessment pending.

0 stars

Status and source references
  • Below the main-section minimum: 0 stars; 100 required.
  • Assessment pending; source references alone do not establish an app or integration.

Source references: src/main/ai/chatgpt-auth.ts, tests/ai-providers.test.ts

shuuul/obsidian-pivi

Pivi is an Obsidian desktop plugin that embeds an AI agent for chatting with, searching, editing, and automating work across a vault.

AI Usage: Unknown; AI usage is not independently verified.

22 stars

Status and source references
  • Below the main-section minimum: 22 stars; 100 required.
  • ai_usage: unresolved. The evidence documents ChatGPT OAuth, vault-context chat, and separate Codex image generation, but the target specification explicitly says live ChatGPT sign-in and chat-turn verification remained pending. Current real-world use therefore is not independently established.
  • integration: contradicted. The claim says the ChatGPT plan integration is confirmed, but the target's own handoff states that live Obsidian sign-in and chat-turn checks were still remaining. Static provider and loader work is evidenced, not confirmed usable runtime integration.
  • availability: unresolved. Version 0.31.0 is a released artifact with downloadable assets, but availability of the ChatGPT integration as a usable app feature is not fully established because live sign-in and chat verification were pending.
  • chain:authorization: unresolved. The target statically defines ChatGPT OAuth, a provider, and a bundled loader, but live authorization was not verified and the specification explicitly lists the live sign-in check as remaining.
  • chain:inference: unresolved. Generic chat routing from input through PiChatService and PiChatRuntime is documented, and a ChatGPT chat turn is specified as a manual test, but the target evidence says that live ChatGPT sign-in and chat execution remained unverified.
  • Confidence: medium. The 0.31.0 source and release package document ChatGPT subscription OAuth, the bundled provider loader, and the released Obsidian artifact, but the project record still lists live sign-in and chat verification as pending.

Source references: specs/056-pi-099-upgrade-and-pi-mcp-client.md

solpbc/solstone-journal

navigate life intelligently

AI Usage: Unknown; assessment pending.

27 stars

Status and source references
  • Below the main-section minimum: 27 stars; 100 required.
  • Assessment pending; source references alone do not establish an app or integration.

Source references: core/crates/solstone-core-chatgpt-auth/src/authorize.rs, core/crates/solstone-core-chatgpt-auth/src/full_tests.rs, core/crates/solstone-core-chatgpt-auth/src/store.rs, core/crates/solstone-core-convey-shell/src/thinking_chatgpt_route_tests.rs, core/crates/solstone-core/tests/chatgpt_plan.rs

spikonado/sprocket

The best and only AI agent for developing both hardware and software

AI Usage: Unknown; assessment pending.

19 stars

Status and source references
  • Below the main-section minimum: 19 stars; 100 required.
  • Assessment pending; source references alone do not establish an app or integration.

Source references: crates/sprocket-server/src/chatgpt_credentials.rs

stevenlee-oai/openclaw

Your own personal AI assistant. Any OS. Any Platform. The lobster way. 🦞

AI Usage: Unknown; assessment pending.

0 stars

Status and source references
  • Below the main-section minimum: 0 stars; 100 required.
  • Assessment pending; source references alone do not establish an app or integration.

Source references: Unknown; no source file or discussion collected yet.

sxp-studio/subjective-zero

SubjectiveZero is a macOS node editor that uses agents to create and live-edit visual effects.

AI Usage: ChatGPT turns natural-language effect ideas into editable node graphs and code, while SubjectiveZero connects an eligible ChatGPT Plus or Pro account through OAuth and uses the resulting plan authorization to run its agent engine. The app also supports model catalogs, reasoning settings, and agent runs through the direct ChatGPT provider.

74 stars

Status and source references
  • Below the main-section minimum: 74 stars; 100 required.

Source references: SubjectiveZero/Modules/Sources/SZAI/ChatGPT/SZChatGPTAccounts.swift, SubjectiveZero/Modules/Sources/SZAI/ChatGPT/SZChatGPTOAuth.swift, SubjectiveZero/Modules/Tests/SZAITests/SZChatGPTTests.swift

sy220284/777

Unknown; no repository description or verified summary is available.

AI Usage: Unknown; assessment pending.

0 stars

Status and source references
  • Below the main-section minimum: 0 stars; 100 required.
  • Assessment pending; source references alone do not establish an app or integration.

Source references: app/src/main/java/com/labteto/dshmobile/local/model/chatgpt/ChatGptAuthCoordinator.kt, app/src/main/java/com/labteto/dshmobile/local/model/chatgpt/ChatGptModels.kt, app/src/test/java/com/labteto/dshmobile/local/model/chatgpt/ChatGptOAuthCallbackServerTest.kt, docs/SECURITY.md

t3snake/gocode

Build my own cli coding agent

AI Usage: Unknown; assessment pending.

1 stars

Status and source references
  • Below the main-section minimum: 1 stars; 100 required.
  • Assessment pending; source references alone do not establish an app or integration.

Source references: src/openaisiwc/auth.go

tashtit/cockpit

Unified desktop hub for coding-agent sessions: Claude Code, Codex and GitHub Copilot, plus Gemini CLI, Cursor, Cline, Roo Code, opencode and Antigravity

AI Usage: Unknown; assessment pending.

2 stars

Status and source references
  • Below the main-section minimum: 2 stars; 100 required.
  • Assessment pending; source references alone do not establish an app or integration.

Source references: Unknown; no source file or discussion collected yet.

theabbie/webiome

Browser-local agent workspace for files, docs, media processing and more

AI Usage: Unknown; assessment pending.

0 stars

Status and source references
  • Below the main-section minimum: 0 stars; 100 required.
  • Assessment pending; source references alone do not establish an app or integration.

Source references: src/worker.js, src/worker.template.js

tutur3u/platform

Tuturuuu is an AI-native, open-source workspace that unifies tasks, scheduling, and team collaboration into a single high-performance ecosystem. Built with Next.js, Flutter, and Supabase, it delivers a seamless and real-time experience across web and mobile platforms for modern teams.

AI Usage: Unknown; assessment pending.

55 stars

Status and source references
  • Below the main-section minimum: 55 stars; 100 required.
  • Assessment pending; source references alone do not establish an app or integration.

Source references: apps/web/src/app/api/v1/users/chatgpt/route.ts, apps/web/src/lib/chatgpt/connections-route.test.ts, apps/web/src/lib/chatgpt/oauth.test.ts, apps/web/src/lib/chatgpt/oauth.ts, apps/web/src/lib/chatgpt/provider-sdk.test.ts

vastsa/PI-Desktop

Local-first AI coding agent desktop: Electron + Rust host core + pi Agent Harness + user-installable plugins

AI Usage: Unknown; assessment pending.

6265 stars

Status and source references
  • Assessment pending; source references alone do not establish an app or integration.

Source references: apps/desktop/test/oauth-standalone-bundle.test.mjs

xibbon/PiSwift

PiSwift is a Swift library for embedding the pi agent and coding-agent stack in Apple applications. Its released v0.87.1 target includes a ChatGPT Plus/Pro-compatible OpenAI Codex subscription OAuth flow, including browser login, token exchange, refresh, and API-key resolution. The evidence supports a usable released authentication integration, but not a fully verified end-to-end inference path.

AI Usage: Planned or incomplete integration: The integration is probable rather than confirmed end-to-end: the released target registers an OpenAI Codex provider, exposes login and refresh functions, and resolves OAuth credentials to an API key, but the supplied files do not show the subsequent model-inference call wiring.

92 stars

Status and source references
  • Below the main-section minimum: 92 stars; 100 required.
  • install: unresolved. A pinned release URL and empty asset list are evidenced, but no release-target package manifest or explicit source-build/install procedure is supplied.
  • setup: contradicted. The rendered setup claim requires a device UUID and describes the ChatGPT-specific flow, but the recommended v0.87.1 target's OAuthLoginCallbacks has no device-ID field and its OpenAI Codex login does not require one; the UUID requirement belongs to the newer ChatGPT-specific implementation at a different commit.
  • requirements: contradicted. Swift/source integration, a host application, Network/CryptoKit considerations, and an eligible ChatGPT subscription are broadly supported, but the rendered requirement of a UUID device identifier is not required by the recommended v0.87.1 Codex flow and is taken from a newer target.
  • chain:inference: unresolved. The evidence establishes credential refresh and API-key resolution but does not show the final provider request or model-stream dispatch consuming the resolved key.
  • chain:distribution: unresolved. A published release tag is confirmed, but its assets array is empty and the evidence does not establish a package manifest, binary artifact, checksum, or documented pinned source-install procedure.
  • Missing required evidence: A release-target package manifest or explicit Swift Package Manager installation instructions.
  • Missing required evidence: A release-target end-to-end test proving that the resolved ChatGPT/Codex credential is consumed by an inference request.
  • Missing required evidence: A packaged binary or checksum for v0.87.1.
  • Confidence: medium. The released v0.87.1 target contains an OpenAI Codex OAuth login, refresh, credential-to-API-key resolution, and provider registration path, but the supplied evidence does not show the final model-inference request wiring or a packaged binary.

Source references: Sources/PiSwiftAI/OpenAIChatGPTOAuth.swift, Tests/PiSwiftAITests/A5OAuthFlowTests.swift, Tests/PiSwiftCodingAgentTests/A5AuthStorageTests.swift

xuzhougeng/wisp-science

Open-source, local-first desktop AI research workbench for scientific computing with Python/R, MCP bioinformatics tools, SSH/WSL/GPU runtimes, and OpenAI/Anthropic models.

AI Usage: Unknown; assessment pending.

998 stars

Status and source references
  • Assessment pending; source references alone do not establish an app or integration.

Source references: crates/wisp-llm/src/chatgpt_auth.rs

xxww0098/dsh-plugin-oauth-subs

dsh-plugin-oauth-subs is a DeepSeek Harness plugin that lets users sign in with ChatGPT, share an eligible ChatGPT plan through the official Sign in with ChatGPT flow, and use ChatGPT Responses models through a local OpenAI-compatible route.

AI Usage: The implementation maps the Harness route to `openai-responses`, uses `/chatgpt/v1/responses`, forwards `Authorization: Bearer <access_token>` to `api.openai.com/v1/responses`, and the end-to-end test verifies a streamed completed response.

8 stars

Status and source references
  • Below the main-section minimum: 8 stars; 100 required.
  • Missing required evidence: No independent runtime service validation beyond the repository's supplied tests and implementation evidence.
  • Missing required evidence: The evidence does not establish operating-system-specific support beyond the documented web and Electron desktop profiles.

Source references: lib/oauth/chatgpt/index.d.ts, lib/oauth/chatgpt/index.js, src/oauth/chatgpt/README.md, src/oauth/chatgpt/index.ts, test/chatgpt.test.ts

zegadb/studio

appbuilder: describe an app, sketch it, and build it as a real desktop app with deka (Tauri); ChatGPT sign-in

AI Usage: Unknown; assessment pending.

0 stars

Status and source references
  • Below the main-section minimum: 0 stars; 100 required.
  • Assessment pending; source references alone do not establish an app or integration.

Source references: src-tauri/src/chatgpt.rs

Unprocessed and pending updates

199 public repositories with queued work; 268 pending stages. Sorted by stars, highest first; unknown stars appear last. Includes unprocessed discoveries and refreshes of previously reviewed repositories. This section shows the full visible backlog regardless of the filters above. Processing also respects queue age, readiness and retry backoff.

RepositoryStarsPending work
openclaw/openclaw391,194AI assessment / verification; Evidence collection / refresh
NousResearch/hermes-agent250,776Evidence collection / refresh
anomalyco/opencode211,504Evidence collection / refresh
earendil-works/pi111,778AI assessment / verification; Evidence collection / refresh
cline/cline69,736AI assessment / verification; Evidence collection / refresh
alibaba/open-code-review43,352Evidence collection / refresh
gitroomhq/postiz-app36,637Evidence collection / refresh
can1357/oh-my-pi34,117Evidence collection / refresh
pingdotgg/t3code24,328AI assessment / verification; Evidence collection / refresh
BuilderIO/agent-native7,038AI assessment / verification; Evidence collection / refresh
vastsa/PI-Desktop6,265Evidence collection / refresh
dagucloud/dagu4,243Evidence collection / refresh
milind-soni/OpenMausBot3,963AI assessment / verification; Evidence collection / refresh
ccch1mneyyy/dsh-TUI3,947AI assessment / verification; Evidence collection / refresh
TanStack/ai3,160Evidence collection / refresh
neavo/LinguaGacha2,527AI assessment / verification; Evidence collection / refresh
K-Dense-AI/k-dense-byok1,293Evidence collection / refresh
xuzhougeng/wisp-science998AI assessment / verification; Evidence collection / refresh
rcarmo/piclaw867Evidence collection / refresh
bastani-inc/atomic843AI assessment / verification; Evidence collection / refresh
kitknox/rootshell732AI assessment / verification; Evidence collection / refresh
narumiruna/pi-extensions647AI assessment / verification; Evidence collection / refresh
ogulcancelik/pi-extensions553Evidence collection / refresh
code-yeongyu/senpi461AI assessment / verification; Evidence collection / refresh
riffado/riffado392Evidence collection / refresh
justrach/codegraff281AI assessment / verification; Evidence collection / refresh
CodeAlta/CodeAlta227Evidence collection / refresh
swarajbachu/zuse219Evidence collection / refresh
orbi-build/orbi194AI assessment / verification; Evidence collection / refresh
miuuyy/Clew187AI assessment / verification; Evidence collection / refresh
Bino5150/lumina181Evidence collection / refresh
ldbumble/taskuary135AI assessment / verification; Evidence collection / refresh
converge-ai-labs/agent-foundation132AI assessment / verification; Evidence collection / refresh
Mintplex-Labs/anythingllm-mobile116AI assessment / verification; Evidence collection / refresh
xibbon/PiSwift92AI assessment / verification; Evidence collection / refresh
AizenvoltPrime/damocles75Evidence collection / refresh
planetaryescape/mxr73Evidence collection / refresh
Demonbane18/relmio56AI assessment / verification; Evidence collection / refresh
tutur3u/platform55Evidence collection / refresh
unrefined803/RPGraph46AI assessment / verification; Evidence collection / refresh
Community-Access/quill45AI assessment / verification; Evidence collection / refresh
pantoniou/fyai42AI assessment / verification; Evidence collection / refresh
tuo-lei/vibe-replay38AI assessment / verification; Evidence collection / refresh
diegopetrucci/the-last-harness36Evidence collection / refresh
leynier/alera36AI assessment / verification; Evidence collection / refresh
arturict/tagvico-ai35AI assessment / verification; Evidence collection / refresh
Sanuu7/AndroidHarness35Evidence collection / refresh
jvm/pi-mono28Evidence collection / refresh
solpbc/solstone-journal27AI assessment / verification; Evidence collection / refresh
ldclabs/anda-bot26AI assessment / verification
fuga-if/idol-live-db23AI assessment / verification; Evidence collection / refresh
maxprain12/dome23Evidence collection / refresh
openclaw-easy/openclaw-easy-desktop22Evidence collection / refresh
kotuku-group/kotuku21AI assessment / verification; Evidence collection / refresh
AmitTzah/ahkllm20Evidence collection / refresh
ManoloRemiddi/augmentor-agent19AI assessment / verification; Evidence collection / refresh
spikonado/sprocket19AI assessment / verification; Evidence collection / refresh
racetozero/kiss18AI assessment / verification; Evidence collection / refresh
gitmaan/stellaris-companion17AI assessment / verification; Evidence collection / refresh
everruns/yolop16AI assessment / verification; Evidence collection / refresh
Joaov41/Rss16Evidence collection / refresh
MOLAaaaaaaa/MolaGPT.Desktop15Evidence collection / refresh
VibeCodyH/pokebench14Evidence collection / refresh
anshuman-pandey/open-granola10Evidence collection / refresh
beeblastco/broods10Evidence collection / refresh
gethamster/horde10AI assessment / verification; Evidence collection / refresh
longyijdos/kana8AI assessment / verification; Evidence collection / refresh
notshekhar/loop8AI assessment / verification
xxww0098/dsh-plugin-oauth-subs8AI assessment / verification; Evidence collection / refresh
draht-dev/draht7Evidence collection / refresh
klovr-co/hover-tag7Evidence collection / refresh
leynier/exeora7Evidence collection / refresh
pejmanjohn/chickpea7Evidence collection / refresh
rcarmo/go-ai7AI assessment / verification; Evidence collection / refresh
lystar-team/lystar-code6AI assessment / verification; Evidence collection / refresh
eaonlabs/eaon-code5AI assessment / verification; Evidence collection / refresh
instrument-org/instrument5Evidence collection / refresh
nimiplatform/nimi5AI assessment / verification; Evidence collection / refresh
luckydye/vektor4AI assessment / verification; Evidence collection / refresh
OrdalieTech/orb4Evidence collection / refresh
depollsoft/MonkeySSH3AI assessment / verification; Evidence collection / refresh
justrach/harness3Evidence collection / refresh
KvFxKaido/Push3AI assessment / verification
masuidrive/pdh3AI assessment / verification
microsoft/amplifier-module-provider-openai-chatgpt3AI assessment / verification
Nitestack/aix3Evidence collection / refresh
pwrdrvr/PwrGit3Evidence collection / refresh
rcarmo/swift-ai3AI assessment / verification; Evidence collection / refresh
Saikrishna-Ambeti/kairo3Evidence collection / refresh
soaringjerry/PCAS3AI assessment / verification; Evidence collection / refresh
chetto1983/Aura2Evidence collection / refresh
IceblueSakura/OpenBridge2Evidence collection / refresh
inclunet/assistente2AI assessment / verification; Evidence collection / refresh
KyrosHendrix/pi-KillerOS2AI assessment / verification
matdev83/go-llm-interactive-proxy2AI assessment / verification
nilsonwhang3-spec/risu-hina2AI assessment / verification; Evidence collection / refresh
rcarmo/rs-ai2AI assessment / verification; Evidence collection / refresh
rubenfb23/research-pi2Evidence collection / refresh
sequelcore/tesota2Evidence collection / refresh
sethhen/t3code2Evidence collection / refresh
tashtit/cockpit2Evidence collection / refresh
aditnotfound/veil1AI assessment / verification
Blue-B/pi-custom-packages1Evidence collection / refresh
calmmage/botspot1Evidence collection / refresh
ephes/pipy1AI assessment / verification; Evidence collection / refresh
eunsoogi/orot1Evidence collection / refresh
ferraridavide/decisionwise1AI assessment / verification
Francigofree/r2b41Evidence collection / refresh
gannonh/kata-code1AI assessment / verification; Evidence collection / refresh
ggfincke/456code1AI assessment / verification
gi0baro/pidrei1Evidence collection / refresh
Gizziio/allternit-platform1AI assessment / verification; Evidence collection / refresh
Huruikagi/tanzakoo1Evidence collection / refresh
jckail/point_bot1Evidence collection / refresh
JiaJunDeng5930/Selvedge1Evidence collection / refresh
LightWhite520/openeden1AI assessment / verification
llms-txt-archive/openai-platform1AI assessment / verification; Evidence collection / refresh
looprig/carbon1Evidence collection / refresh
lse872317312010/personal-os1Evidence collection / refresh
m-rui001/MATE1Evidence collection / refresh
microsoft/amplifier-unified1AI assessment / verification; Evidence collection / refresh
miziusLabs/Dray1Evidence collection / refresh
nourdean22/MAINnicks-tire-autoNEW1AI assessment / verification
Suraj-H675/Ash-Harness1AI assessment / verification; Evidence collection / refresh
t3snake/gocode1Evidence collection / refresh
travnie/aistee1AI assessment / verification; Evidence collection / refresh
trustdan/acctg-practice1Evidence collection / refresh
AdilShamim8/ReWoo0Evidence collection / refresh
aeltawela/chatgpt-as-provider0Evidence collection / refresh
akshatporwal002/Memory0AI assessment / verification
Anemos-labs/Sleipnir0AI assessment / verification; Evidence collection / refresh
AojdevStudio/nzube0AI assessment / verification
attrition-tech/arkex0Evidence collection / refresh
aydinmrnv/relay0Evidence collection / refresh
CaT-Hode/ASS-Agent-Switching-Servant0AI assessment / verification
Chia1104/solyx0Evidence collection / refresh
coji/asist0Evidence collection / refresh
CrowBe/weave0Evidence collection / refresh
Cyb3rb1ade/PLUR1BUS-Harness0AI assessment / verification; Evidence collection / refresh
daydaylx/nixpi-core0AI assessment / verification
detroitpro/t3code0Evidence collection / refresh
domehahn/harnessmesh0AI assessment / verification
dotdioscorea/eyeballs0Evidence collection / refresh
edlundin/kiln0Evidence collection / refresh
Elijah-J/chatgpt-plan-sdk0Evidence collection / refresh
epicycloids/multivac-client0Evidence collection / refresh
Fallstop/codeslop0AI assessment / verification
GodStorm91/TrainOrRest0AI assessment / verification
hafiezul/rove0AI assessment / verification
isaachorowitz/TapLyne-App0AI assessment / verification
jamestrew/pi-bites0AI assessment / verification; Evidence collection / refresh
JoshAaronLevy/collie-writer0Evidence collection / refresh
kineto-app/kinetik0Evidence collection / refresh
kkestell/ox0AI assessment / verification
kongdd/ai-quota0Evidence collection / refresh
ksgat/zither0AI assessment / verification; Evidence collection / refresh
Kudl1k/AudioNotes0AI assessment / verification
kzinmr/ai-topics-pi0AI assessment / verification
laurensvd87/universal-discussion0AI assessment / verification
leandromesq/lelezonio-pi-kit0Evidence collection / refresh
leonpresistforever-png/Oceanstudio.apk0Evidence collection / refresh
LisaScheers/Forge0Evidence collection / refresh
logno-dev/ocr-list0AI assessment / verification
MelanieMende/FinPal0Evidence collection / refresh
mhismail3/tron0Evidence collection / refresh
Mik-pe/Knut0AI assessment / verification; Evidence collection / refresh
misaon/workbox0Evidence collection / refresh
MYLEGND/masterapp0Evidence collection / refresh
Nana1873/stardew-i18n-translator0AI assessment / verification
NateWeav/t3code-hermes0AI assessment / verification
neuroforge-io/Sinter0AI assessment / verification
ngpestelos-mirrors/openclaw0AI assessment / verification; Evidence collection / refresh
ntindle/don-a-token0Evidence collection / refresh
Ogekuri/PI-useReq0Evidence collection / refresh
olidepalma/GarminTrainingPlaner0Evidence collection / refresh
Owlbay/armadra-agent0Evidence collection / refresh
OxFrancesco/Labora0Evidence collection / refresh
Qin-long/sub2api0Evidence collection / refresh
Rasalas/tau0AI assessment / verification; Evidence collection / refresh
rmonvfer/micro0Evidence collection / refresh
ronen/postcode0Evidence collection / refresh
sakompella/data-dump0AI assessment / verification
SergeSerb2/t3-pretty0AI assessment / verification; Evidence collection / refresh
servoagents/entryplug0AI assessment / verification
sgttomas/chirality0Evidence collection / refresh
ShadyAwad/Horizon-HR-v20Evidence collection / refresh
shriism/serenity0Evidence collection / refresh
Srimi1/antigravity--mobile0Evidence collection / refresh
StefanDG1/vibe-scroller0AI assessment / verification; Evidence collection / refresh
stevenlee-oai/openclaw0Evidence collection / refresh
sy220284/7770AI assessment / verification; Evidence collection / refresh
TamynatorSama/JobPrep0AI assessment / verification; Evidence collection / refresh
Taskov1ch/mecha-maiden0AI assessment / verification
TGoddessana/alpine-code0AI assessment / verification; Evidence collection / refresh
theabbie/webiome0Evidence collection / refresh
triangle-int/nolune0AI assessment / verification
tripplyons/pi-extensions0AI assessment / verification; Evidence collection / refresh
UnknownAlienHuman/eliot-swarm-controller0Evidence collection / refresh
zegadb/studio0Evidence collection / refresh