Clasificación de plugins

Uptimely

Uptime monitors and incidents

Gondwana

MICHAEL LEE ADKINS

Build, debug, and understand games made with the Gondwana C#/.NET game engine. The plugin uses Gondwana-specific workflows and read-only access to the current official repository, tests, demos, templates, and wiki so answers and generated code can be grounded in the engine's current public APIs.

Archkeel

Rapiddweller

Use the Archkeel CLI to inspect a repository, review architecture contracts, and check coding-agent changes. Python static analysis and Dart/TypeScript import graphs have explicit limits; missing evidence remains UNKNOWN. Install the CLI separately; its Python package includes all three language collectors.

QueryHost

Check and compare game servers

Hopper

Hopper

Hopper is a realtime inference API for open-source models: two LLMs, text-to-speech and speech-to-text, behind a standard chat-completions API. Try the latest open-source models on your own work: in a chat, Hopper runs your prompt or examples on both LLMs side by side and returns each answer with time to first token, tokens per second, token counts and cost, so you can judge the models on your task. It also speaks text with an inline audio player, transcribes recordings you attach (WAV, MP3, M4A, Ogg, FLAC or WebM) with word timestamps, lists models and prices, and returns tested client code for popular SDKs and agent frameworks. After you sign in with your Hopper account, it shows your usage and creates API keys: the key opens once in the Hopper console, with a ready snippet for your code. Where it can run commands in your project, the plugin also finds the prompts your app sends and tests them on both models, and after you say go it points your project's chat-completions client at a Hopper model, changing only the client. Text and audio you send go to Hopper's API; see withhopper.com/privacy.

Codiedev Reports

Codiedev

Codiedev Reports answers engineering leaders' questions from their Codiedev reporting, right in the conversation. Ask how engineering is doing this month, whether AI-assisted pull requests actually merge faster, which repositories drive AI spend, how adoption is going on a team, or which models people use and how reliable they are. Who it's for: company admins, CTOs and team managers whose company uses Codiedev to capture AI coding sessions and pull requests. Admins see the whole company and can narrow to any team. Team managers see only their own team and never see spend. Engineers can't connect. What it can do: read delivery and cycle-time stages, the AI-vs-no-AI comparison with its classification coverage, velocity and the throughput forecast, daily activity and rate-limit pressure, adoption and skill reuse, model usage and reliability, capture coverage, and (admins only) spend versus the prior period, ROI, cost per merge, seats and the spend forecast. Limitations: it is read-only and can't change anything in Codiedev. It reports team and company aggregates, not individual session transcripts. Numbers come from stored rollups refreshed about hourly, over 30, 60 or 90-day windows, and only cover sessions Codiedev captured; every answer can quote capture coverage so you know how complete it is. A Codiedev account with the Admin or Team manager role is required.

SHIP

Agentic coding for your SDLC

FiraQuantumSafe

Fira Software Ltd

Check a public website's HTTPS encryption and post-quantum TLS with measured evidence. FiraQuantumSafe helps website owners, developers and security teams inspect TLS 1.2 and TLS 1.3, certificate trust and expiry, security-header presence, and hybrid X25519MLKEM768 key exchange. Retrieve reports using their IDs and compare observations for the same hostname after configuration changes. Each report describes the tested public HTTPS endpoint, and unavailable measurements remain unknown. Public reports remain available for 30 days. You can optionally keep an exact report in your account on the FiraQuantumSafe website.

Autonom

Andrei Iatsuk

Autonom gives AI coding agents a portable skill pack and JSON CLI to test and debug mobile apps on Android and the iOS Simulator: explicit target sessions, compact accessibility/UI trees, semantic taps and gestures, screenshots, recordings, logs, crash reports, deep links, permissions, app-container files, a consent-gated localhost MITM proxy for inspecting, mocking, and exporting HTTP(S) traffic, and Flow v1 — repeatable flow files with exact selectors, polling assertions, and machine-readable failure classes. Requires a separately installed Autonom CLI, Python 3.11+, and the device tools for your platform. The iOS Simulator requires macOS and Xcode.

HelpTester

HelpTester

HelpTester brings its browser-based hardware tests and developer tools into ChatGPT. Interactive hardware tests that run directly in the chat: Network checks: Developer tools: Tests that need the full browser (webcam, dead pixel, CPU/GPU stress test, favicon checker/generator, tag checker, song key finder) open on helptester.com. All tests use standard browser APIs and run client-side. No AI is used for the tests themselves, nothing is recorded or uploaded unless you explicitly save a result.

Goalmatic

Goalmatic

Create website and simple App projects from a brief, inspect source files, make revision-checked edits, and open previews in your Goalmatic workspace. Draft previews can be accessible to anyone with the link; unpublished does not mean private. Publish an exact project version after approving it on Goalmatic, then check deployment status. A Goalmatic account and connection are required. You choose the workspaces and permissions to grant. This plugin supports project source and publishing; it does not send messages, process payments, install Apps, list Apps in the Goalmatic App Store, or execute arbitrary platform operations. Stored quality results do not replace testing your project.

mailX

Mailwarm Inc.

mailX finds out why your emails land in spam or junk, using live DNS checks, and tells you exactly what to fix. Ask in plain English, for example "why are my emails going to spam?", "does my domain have the SPF, DKIM and DMARC that Gmail and Yahoo require for bulk senders?" or "check DMARC for acme.com". mailX runs the checks in real time: Useful when you set up a new domain, move to Google Workspace or Microsoft 365, launch a newsletter or an outreach campaign, or see emails going to spam in Gmail or Outlook. Every result comes from a live DNS query, so you see the state of your domain right now. No account or setup needed. mailX only reads public DNS data. It never changes your DNS, never sends email and never asks for passwords. Built by Mailwarm Inc. (YC S20), an email deliverability company.

camelRun

camelAI

camelRun hosts persistent AI agents for developers and teams. Describe an agent's job, give it instructions and supported tools, and run it from the conversation. Agents keep their conversation history until deleted and can run scheduled tasks or pause for a person's answer or approval. Create a research assistant with web search, continue an existing agent's work, review its replies and tool activity, adjust its model or instructions, and manage its schedules. For reusable configuration, deploy agent.yaml manifests with versioned definitions and preview changes before applying them. The plugin operates only on the camelRun account you connect. Runs use that account's credit. Agents can use built-in tools and configured integrations; applications serving attached tools must be connected for those tools to work. Manage billing and provider keys in the camelRun console. Deploy tool connections requiring secrets with the camelrun CLI so credentials stay out of chat. Deleting an agent or definition is permanent and requires confirmation.

reverse engineering anything

Evidence-led REA workflows

Ultralytics YOLO

Train and deploy vision AI

TALON Knowledge

Search TALON manuals, Q&A

Cheapest Domains

ILIOS GALIL

Focus on the real cost of keeping a domain: its annual renewal price. Compare selected registrars, with initial registration prices shown separately and supported multi-year estimates. Your assistant suggests names and checks availability and premium status. Preserve source dates, partial coverage and unknown results. Renewal rates can change; purchases and exact premium prices remain at the registrar. Optional Cloudflare Radar data measures DNS popularity, not Google search competition. The native form collects seed names and an optional project brief; your current assistant generates and checks suggestions. The website form prepares a request to copy into a connected assistant. Website subscription authentication is not implemented.

MA Commander

FABIO ANDREOLA

MA Commander lets users securely inspect and manage an authorized remote computer through MCP. It supports files, processes, developer commands, persistent sessions, WSL workspaces, resumable transfers, screenshots, policy-gated GUI control, and limited privileged operations. Access is restricted by MA Commander safety policy and the permissions granted during authorization.

Pipecat Context Hub

Build Voice Agents: Pipecat

CyberLens AI

Jeramy Adam Ratelle

Connect your CyberLens AI account to review recent website and repository scans you own. Ask for recorded scores, scan dates, status, and test counts when available. The plugin provides scan summaries, not finding evidence or a current security guarantee. It reads saved scans and cannot start scans or change your account.

Aevral

Better ISMS

Aevral is an AI security reviewer for GitHub pull requests. Each review looks for access control, business logic, injection, XSS, SSRF, path traversal, unsafe deserialization, token and session flaws, and LLM-integration risks, and posts each finding with a suggested fix you apply. This plugin is a setup skill: it walks you through installing the Aevral GitHub App so reviews run server-side. It does not review code on your machine. After the App is installed, the next pull request can already be reviewed, including before anyone signs in to the console. Reviews are advisory and never block a merge. The optional whole-repo scan reads authorization, IDOR, and business-logic access control only; it is not secret, dependency, or memory-corruption scanning, and not a general SAST. GitHub only. By Better ISMS.

CodeAgent Mobile

CodeAgent Mobile

Describe an app and CodeAgent builds it: a live web app, a shared app or a native phone app, ready to use and share by link. Before anything starts you see what will be built and confirm it. The build card shows its progress until the app is ready to open. You can also search CodeAgent Remix, a feed where every post is a live app, for apps other people made, and remix any of them by describing a change. For developers, the same connection supervises the coding agents linked to your CodeAgent account in dedicated sessions: start a task or send a new instruction, answer the questions the agent asks, stop it, read the changes it made and open a pull request. Hand a session to your phone when you step away and keep going from the CodeAgent Mobile app. Building or remixing an app never starts without your confirmation, and switching a session to another agent always asks first. Building apps needs only a CodeAgent account. The coding-agent tools need at least one paired session (local CLI, GitHub Codespace, CodeAgent Box or self-hosted server).

deepface.dev

Tech Local

Build facial verification into your app without hosting machine-learning models. This plugin helps your coding agent connect the deepface.dev API, compare two face photos, generate face embeddings and handle verification results in your existing stack. It guides account signup and server-side key setup, then tests the integration with images you are allowed to process. An optional brief project summary helps the deepface.dev team understand your needs; private code, conversations and customer records are not requested. This is an integration skill, not a connected image-analysis tool. A deepface.dev account is needed to call the hosted API. The account owner confirms their email and accepts the service terms. Facial similarity is not proof of identity or liveness.

Lockbox

Opslane

Send API keys and private files to a teammate's agent without pasting them into chat. How it works in ChatGPT: The assistant never sees your file or secret through Lockbox. A file reaches the chat only if you click "Send to chat". Lockbox's servers store only encrypted data.