插件

插件排行榜

按采集日期查看 ChatGPT 公开插件目录的展示顺序。热门分类是官方精选,目录未公布安装量或使用量。

最近采集

#577
ChatGPT

Agent Signals

ZazenCodes

Agent Signals by ZazenCodes tracks what is changing in AI coding. Every update cites an official source (changelogs, release feeds, docs, model pages, vendor announcements) with a verbatim evidence quote. News that a trusted practitioner reports first is labelled as reported until an official source confirms it, and practitioners' posts are attached as attributed commentary. Ask what changed in Codex this week, which coding models dropped this month, what practitioners think of the latest Claude Code release, or for a weekly briefing to share with your team. Every answer shows how fresh the data is.

#578
ChatGPT

Uniquel UI Components

Uniquel

Use this when Codex is writing code in a Next.js or React app styled with Tailwind CSS. Search the Uniquel catalog and insert the source for a component or page section into the project. Do not use it in a chat with no codebase, or for design feedback, mockups, and layout ideas.

#579
ChatGPT

Vibe App Scanner

Silex Dev

Review web-app security findings without leaving your conversation. Connect an eligible Vibe App Scanner account to see recent scans, check progress and read available reports with remediation guidance. You can also start a standard scan of a public app after confirming your ownership or permission to test it and accepting the VAS scanning terms. Scans use your existing VAS allowance, and report access follows your account permissions. VAS does not modify your application, accept login credentials for target sites or run recurring scans through this plugin. Reports omit raw evidence and source files. Findings are observations, not a guarantee of security; failed or incomplete checks remain unknown. A VAS account with plugin access is required.

#580
ChatGPT

Email Domain Doctor

Ahmad Khodr

Email Domain Doctor inspects the public email DNS setup of any domain. It validates MX records, checks SPF configuration and mechanism policies, parses DMARC policies and reporting addresses, and optionally inspects DKIM records when a selector is provided.

#581
ChatGPT

Currents

Currents Software Inc

Currents gives AI agents the test results your CI records, from Playwright, Vitest, Jest and more. Self-healing tests• Ask why a run failed and get the errors, stack traces, screenshots and traces behind it• Step through a run's Playwright trace, action by action, to see what the page did• Find the root cause and fix the test, flaky ones included• Compare a pull request's runs with main to see what the change broke Proof of work for any change• Before-and-after screenshots, videos, Playwright traces and DOM snapshots from CI• A demo the agent records in a browser, saved as a run• Evidence links posted in the pull request, so reviewers see the change working Test suite health• Rank your flakiest and slowest tests and spec files• See which errors break the most tests• Quarantine, skip or tag tests with rules, and review what they affected Day-to-day CI work• Reproduce a bug from a failed run and verify the fix• Open an issue for a failure in your issue tracker• Cancel, reset or re-check runs without leaving the chat For teams whose CI slows them down: flaky tests that block merges, failures nobody has time to dig into, and AI-written pull requests that are hard to trust without seeing the change work. Works with GitHub, GitLab, Slack, Jira, Linear and more. You need a Currents account.

#582
ChatGPT

KamuiDash

KamuiDash

KamuiDash helps you manage the apps you own on KamuiDash. Inspect projects, apps, deployment status, and logs. With your explicit confirmation, create and update apps, create databases, or configure custom domains.

#583
ChatGPT

AgentDust

Hamza Hamidi

Explain AgentDust process and disk reports that users provide, and guide local setup. This skills package does not connect to a Mac or provide live process access. Live inspection requires a separately configured local AgentDust MCP connection. Cleanup requires AgentDust's own typed approval.

#584
ChatGPT

AppTail

AppTail

AppTail helps iOS developers decide what to build, where to launch, and how to improve App Store visibility. Research niches across storefronts, compare competing apps and their ratings, explore keyword popularity and search results, and turn listing and review evidence into practical ASO recommendations. Works for pre-launch ideas as well as existing apps. Connect your AppTail account with OAuth. Data access and tracking limits depend on your plan. Public app data and competitor estimates are distinct from measured App Store Connect analytics for your connected apps. Competitor downloads and revenue, when available, are rolling 30-day estimates. AppTail does not expose a standalone numeric keyword difficulty score, cover Google Play research, or publish listing metadata. Recommendations are based on available data and do not guarantee rankings or downloads.

#585
ChatGPT

VARUA Remote Control

VARUA

Connect ChatGPT and other MCP clients to computers you explicitly pair with VARUA Remote Control. Inspect and search files within locally authorized roots, review processes, and—when the machine owner enables them—write files, run bounded commands, and use persistent terminal sessions. Remote Control routes requests through outbound authenticated agents while mutation and execution permissions remain under local device control. Authorized filesystem roots constrain dedicated filesystem tools and command working directories; an approved command or terminal runs with the local agent user's OS permissions and is not an operating-system sandbox.

#586
ChatGPT

PromptBrake Free Tools

Ammar Jirjees

Prepare practical tests and release plans for your AI agents and chatbots. Get prompt-injection test inputs, map OWASP LLM risks to test prompts and warning signals, create an AI release plan with open decisions and a starter gate policy, and build a tests.json pack containing 1–20 response checks for PromptBrake CI. These free tools require no account or API key. They help you prepare tests; they do not run scans against your application. Response checks use case-insensitive text comparisons, and the release-planning score measures planning completeness, not security assurance. Use test inputs only with systems you own or are authorized to test.

#587
ChatGPT

Aqteron

EIREEN Tech

Build and manage web applications in your connected Aqteron account. Describe what you need, then use current platform instructions and specialist guidance to create an application package. Validate a ZIP before publication, publish a new app, update an owned app with version checks, inspect retained versions, retrieve source ZIPs, and check deployment results. Publication makes the chosen version available at its returned web address. An Aqteron account and authorization are required; available features depend on your account.

#588
ChatGPT

ATL Sandbox

VizPageAI

ATL Sandbox gives ChatGPT isolated cloud computers for your account. Ask in plain language and ChatGPT creates a sandbox, runs Python, Node.js, C/C++, Rust, or shell commands, and returns the results to the conversation. What you can do: Who it is for: developers, analysts, students, and creators who want their assistant to run real code and keep the results. Security: every sandbox runs behind its own user-space kernel and can reach only the compute, storage, and database of the signed-in account. Infrastructure credentials never enter the conversation. Actions that change or delete data are marked so ChatGPT asks before running them. Limitations: an ATL Sandbox account is required. Accounts have usage limits on concurrent sandboxes, run time, storage, GPUs, and schedules, and running compute stops when an account reaches its limit. Instant sandboxes discard their files when stopped. Images must be linux/amd64 and include /bin/sh. Interactive terminals and Codex sessions are not available in this app. Do not paste passwords, API keys, or payment details into sandboxes.

#590
ChatGPT

FlutterGo

FlutterGo.AI

FlutterGo.AI connects ChatGPT to a cloud Flutter studio. Sign in with your FlutterGo.AI account, then build a mobile app step by step in the chat. What you can do:

#591
ChatGPT

Deriv

deriv

Live Deriv API schemas, payload validation, and guides for authentication, subscriptions, trades, and errors. Search endpoints, read current request and response fields, use worked examples, and validate payloads. This plugin doesn't sign in to a Deriv account, store keys, or place trades.

#592
ChatGPT

Superlog

Superlog

Connect the Slack channels where errors arrive and the repositories your team works on. Superlog investigates reports, opens bugfix pull requests when it can verify a fix, and replies in the original thread with its findings. Describe what you want to automate, and we’ll walk you through getting started—from creating your Superlog account to connecting your tools and activating your first automation. You can also set up scheduled reliability checks, Sentry issue triage, and support workflows. Connect an existing Superlog account or create one during setup.

#593
ChatGPT

TokenOS dApp Builder

TokenOS

Turn an idea into a buildable dApp spec: detected features (NFT mint, token page, staking, swap, DAO, marketplace, game, dashboard, AI agent…), pages, components, data model, wallet and contract integrations for Solana, Base, Robinhood Chain, Ethereum or Arbitrum, the recommended model and a credit estimate. One link opens the TokenOS Development Workspace with the prompt pre-filled — build, preview, deploy and host without leaving the browser. Tools:• plan_dapp — Turn the user's app idea into a spec (features, pages, components, data model, Web3 integrations, stack, steps, credit estimate) with a pre-filled TokenOS IDE link. Only call with the user's own description — if the request is vague (e.g. “build an app”), ask the user what the app should do and for whom first; never invent an idea.• recommend_models — Pick the best, balanced and cheapest frontier models from the live TokenOS roster for a build task, with price multipliers.• tokenos_capabilities — Reference: supported chains, deployment targets, hosting, token launch, agents, payments and credits. Targets questions like: “build a dapp with ai”; “generate solana app”; “ai web3 app builder”; “create nft minting website”; “build dao website”; “no code dapp builder”. Documentation: https://tokenos.ai/mcp/dapp-builder Outputs are informational — on-chain reads, deterministic calculators and heuristic scans — not financial, legal or security advice. TokenOS is non-custodial and never holds keys or funds; anything you launch, deploy or sign happens from your own wallet.

#594
ChatGPT

Numerical Research Kit

Adem Vessell

Compare baseline and candidate numerical predictions against reference values. Inspect mean squared error, declared costs and fixed-correction bounds, then freeze and verify saved records. Includes deterministic Python tools, strict input checks and a synthetic solver example. Requires Python 3.9 or later and file execution. Costs and references are supplied by the user; no automatic cost measurement, statistical confidence or scientific certification is provided.

#595
ChatGPT

Accessibility Checker WCAG

Brightstep Tech Inc.

Accessibility Checker WCAG checks a public web page for accessibility problems. Give it a URL and it loads the page, and optionally up to 4 more pages linked from it, in a real headless browser, runs the open-source axe-core rules for WCAG 2.2 Level A and AA, and lists what to fix first: each issue's severity, the WCAG success criterion, how many elements fail, example selectors and HTML, and a plain-language fix. It catches problems that need a rendered page, such as low colour contrast, missing alt text, unlabeled form fields, nameless buttons and links, and small touch targets. After you fix things, ask for a rescan and it shows what was fixed, what is new and what changed. Issues inside third-party embeds such as YouTube videos are tagged and listed after your own. It also looks for a link to an accessibility statement (Erklärung zur Barrierefreiheit). Every report adds a short checklist of what automation cannot test, and context on Germany's BFSG, the European Accessibility Act (EAA) and the ADA, in German or English to match ChatGPT's language. Automated checks find only part of all accessibility issues, so a report is not a legal conformity assessment or certificate. Pages behind a login, intranets and private addresses cannot be scanned.

#596
ChatGPT

upAPI

Devino Solutions

upAPI is an API marketplace: one account and one API key for a catalog of ready-to-use APIs, so a team does not sign up for and wire each service separately. The upAPI plugin brings a focused set of those APIs into ChatGPT. What it can do: Tools: eleven tools (fetch_markdown_post, screenshot_post, html_to_pdf_post, pdf_extract_text_post, image_ocr_post, audio_transcribe_post, audio_transcribe_result_get, github_repo_get, npm_package_get, wikipedia_article_get, currency_convert_get). None of them looks up people, and none asks for personal information. Who it is for: developers and teams who already use upAPI and want those APIs inside a conversation. How it connects: you sign in to upAPI with OAuth 2.1 (PKCE) and approve the connection on upAPI's consent screen. You can revoke the connection at any time from your upAPI settings. Limitations: files and pages are processed for the request and not kept as documents; extracted text is cached for up to 24 hours and transcripts are kept 24 hours so they can be fetched. The plugin does not send email, post anywhere, process payments or generate media.

#597
ChatGPT

Remote Arc

BAIYANG YAO

Remote Arc securely connects ChatGPT and Codex to Windows, macOS, and Linux computers that the user explicitly pairs. It can list paired devices, inspect directories and files, read process information, read explicitly shared browser tabs, run terminal commands when enabled, modify files, manage Remote Arc background processes, and restore supported local file edits. Each device has an independent revocable credential and per-device tool policy. Remote Arc uses an OAuth-protected Universal Remote MCP endpoint and an outbound-only device connection. Use it only with computers, files, accounts, and services you own or are authorized to administer.

#598
ChatGPT

Agent Memory

Rohit Ghumare

Recall past decisions, resume work, save useful context, and learn from corrections across your coding sessions using one local Agent Memory daemon. Includes search, session history, lessons, graph queries, memory maintenance, export, and deletion workflows. Requires Node.js 20+ and a running Agent Memory daemon. Basic keyword recall needs no model key. Embeddings and model-based features require configuration; external providers may receive selected data when enabled. This package contains the skills only. Install the Agent Memory daemon and wire its MCP server with agentmemory connect codex so the skills can call the memory tools. Designed for local Codex sessions.

#599
ChatGPT

WishKit

WishKit LLC

WishKit collects feature requests and votes from your iOS app or website, and imports your App Store reviews. Connect it to ask about that feedback in plain words: which requests have the most votes, which are gaining momentum, what paying users ask for, what people say in their reviews, and what to build next. WishKit is read only here. It can't create, change or delete anything. Answers follow what your WishKit plan shows, and names and emails of the people who asked and voted are shared only if you allow it when connecting.

#600
ChatGPT

PromptingCraft.com

Aleksandr Prokofyev

PromptingCraft.com helps you write, review and improve prompts for coding agents such as Codex, Claude Code and Cursor. Turn a development discussion into clear, execution-ready instructions for an AI developer. It identifies the task contract (requirements, decisions already made, permissions and open questions), checks applicable official guidance from AI developers, writes a draft, critiques it against the contract, and returns one final copyable prompt. Optional ideas stay optional, unresolved choices stay open, and no new permissions are added. It can also review an existing prompt: it points out unnecessary instructions and changes in scope, and when a prompt is already suitable it says so and leaves it unchanged. Bundled model-specific records include official OpenAI and Anthropic guidance and scoped DeepSeek and Kimi API tool-use guidance; Cursor behavior comes from Cursor documentation. A public source registry records where each recommendation comes from, and the project's own methodology is labeled separately from official guidance. The skill works on its own. The optional PromptingCraft.com guidance server (mcp.promptingcraft.com) adds read-only access to its deployed guidance ledger: it lists official guidance for a target agent, model family and topic, returns a record with its source, limits and verification date, and checks whether an official source page changed since it was verified. Its tools take only fixed categories and identifiers as input and do not ask for task text. Checking current guidance depends on the tools and access available in the conversation. When the server or live retrieval is unavailable, PromptingCraft.com says so and states whether it used other official sources, dated cached guidance or none. PromptingCraft.com writes and reviews prompts only. It does not run the development task, edit code or grant permissions, and it does not guarantee better downstream results. It needs no PromptingCraft.com account or credentials; installation and connection availability depend on the host. The guidance tools take only fixed categories and identifiers, and no PromptingCraft.com service receives your prompts. Evaluation methods, results and limits are published at promptingcraft.com/benchmarks. PromptingCraft.com is developed independently and is not affiliated with or endorsed by OpenAI, Anthropic or Cursor. Developed by the dondego.com team.