Lint MCP tool schemas across eight clients and score how well agents recover from errors
Runs locally- Open source
- Developer tools
Lint MCP tool schemas across eight clients and score how well agents recover from errors
Runs locallyGive coding agents a shared team, claimed work and a record of who did and accepted it
Runs locallyRead-only access to your Garmin Connect sleep, HRV, training and recovery data
Runs locallyMCP server that OCRs local scanned PDFs and images and exports bank statements
Runs locallyOpen-source DAW your agent drives over MCP, hearing renders as text
Runs locally
Free Remotion and HyperFrames video templates your agent plans and renders locally
Not checked yetSearch 267 animated React components and install them with shadcn from your AI agent
Not checked yetMCP server that uploads documents to DocuPipe and returns extracted structured JSON
Runs locallyLocal-first memory for Claude Code that scores, fades and consolidates what it keeps
Runs locallyCrypto and RWA perp microstructure data for agents, via Parquet handles
Not checked yetScrape pages and map sites from an agent, processed on EU servers
Runs locallyOperate Proxmox through an AI agent with planned changes, receipts and undo
Runs locallyLocal MCP server that lets agents search and write Markdown notes in your repos
Runs locallySend transactional email and manage domains, templates and webhooks from an AI agent
Not checked yetFind, run and read results from Scrape.ly Actors inside any MCP client
Not checked yetFind end-of-life runtimes, packages and Actions in a repo, with file, line and source
Runs locallyLet an agent trade spot and perpetuals across Solana, EVM chains and Hyperliquid
Runs locallyDeploy and operate self-hosted OpenClaw on AWS, GCP, Azure or any VM via typed tools
Runs locallyRead Opik traces, score them and save prompts from your coding assistant
Not checked yetUnofficial Proton Mail Bridge server: read and draft mail, approve every change
Runs locally
Email inboxes, SMS numbers and workspace accounts for AI agents
Not checked yet
Let an agent drive your logged-in Chrome, then replay the compiled task at zero tokens
Not checked yet
Delegate tasks from Claude.ai to your self-hosted OpenClaw agent with OAuth 2.1
Runs locallyPay-per-call tools for agents: company records, screening, validation, web reading
Not checked yetLocal mail, calendar and tasks for agents, with isolated accounts and guarded writes
Runs locallyLive web search and clean-markdown page fetch for coding agents, keyless by default
Not checked yetShared local Markdown memory for coding agents across Copilot, Cursor and Claude Code
Runs locallyCompare cities on sourced data for take-home pay, rent, visas, climate and safety
Runs locallyScores churn risk from 0 to 100 using sequences of customer messages
Runs locallyAnti-bot-aware scraping: crawl, render, screenshot and unlock pages via one API
Not checked yet