All MCP servers

Create and manage Instagram comment-to-DM keyword automations in CommentKeyword.

Directum RX for AI assistants: assignments, documents, boards, projects, any entity, system help

Read, calculate, edit, and save XLSX workbooks headlessly over local stdio.

Your own IMAP/SMTP mailboxes as agent tools, with per-agent policy and human-approved sends.

Compare cloned repositories against taxonomy packs: one code graph, cited claims, a matrix. Local.

Let coding agents play, test and tune your 2D web game: live state, explained errors, replays, fuzz.

Drive a running Limoni terminal app by its semantic tree: find, click, type, wait_for.

64 CQRS tools exposing 234 GitLab operations across 26 entity types

Deploy and manage apps on your own AppCrane instance, the self-hosted PaaS for AI-built apps.

Aggregate long time-series ranges with a few source requests instead of one per bucket.

Manage cross-repository beads work, hive status, and Beadhive workflows through bh-mcp.

Evidence-backed web research for AI agents with citations and confidence scores.

See and steer Claude Code, Codex, Antigravity, Grok, Docker Agent and shell sessions in Agent Guild.

Mail between coding-agent sessions, on one machine or across machines; relays to a Swarmail server.

Publish pages, documents, or folders to a link; documents render as pages; updates keep the URL.

Look up APIKO image and video models, quote prices, generate media and track spend.

MCP tools + agent skills in one zero-dependency CLI: 71,929 -> 581 tokens (-99.2%, measured).

Three.js MCP server: see, play, test and profile the three.js and WebGL pages your agent builds

Generate buildable Compose Multiplatform and Unity game projects locally from tested templates.

Read, check and write Rowih, a language for AI agents whose moves state force, basis and authority

A local CRM as Markdown files in git: read and update pipeline, contacts and interactions.

Local search over your code and documents for AI agents.

Open huge JSON locally and give agents its shape, queries and exact values, never the whole blob.

Testing MCP for coding agents: exercise real user journeys, find bugs, verify them with evidence.

Local-first Mac meeting memory: open promises, people, next-meeting briefs. Read-only.

Cut LLM agent token use in Claude Code, Cursor, Codex, Gemini CLI. Local, no API key. gAI Ventures.

Dry-run network changes in a real containerlab lab: BGP, routes, reachability, with evidence.

Local-first multilingual, auditable memory for Codex, Claude Code, and MCP clients.