-
TeamBrain MCP server + local daemon: serve team memories to any MCP-capable coding agent
Version check not run yet
Memory and context npm Local 4 a week APACHE-2.0
-
A lightweight time machine for AI coding agents
Version check not run yet
Coding agent tools npm Local 4 a week MIT
-
Portable agent working state MCP server — tasks, plans, notepads, memory, and rules for any AI coding agent
Version check not run yet
Coding agent tools npm Local 4 a week MIT
@M
@tinyport/mcp io.github.muhammadali511/tinyport-engine-mcp 1.5 · 30/100 MCP server that gives AI coding agents TinyPort's import-cost analysis: sizes, unused imports, budgets, and safe fixes.
Version check not run yet
Coding agent tools npm Local 4 a week MIT
-
Stop AI coding agents from installing hallucinated or malicious packages. Verifies a dependency's provenance and reputation before it gets installed — built to catch slopsquatting.
Version check not run yet
Coding agent tools npm Local 4 a week MIT
@C
@triago/cli io.github.priyanshun/triago 1.5 · 30/100 Human-in-the-loop decision surface for CLI coding agents: the agent posts a card of findings, you triage them in a browser with the keyboard, and your decisions come back as structured data. Local-onl
Version check not run yet
Coding agent tools npm Local 4 a week MIT
-
MCP server exposing your True Memory profile to coding agents
Version check not run yet
Coding agent tools npm Local 4 a week MIT
-
A firewall between AI coding agents and dangerous actions. Blocks hallucinated packages, leaked secrets, destructive commands, and test-subversion.
Version check not run yet
Coding agent tools npm Local 4 a week MIT
AL
alteran io.github.darkclaw921/alteran 1.5 · 30/100 alteran — coding agent terminal with a built-in phased task tracker
Version check not run yet
Project management npm Local 4 a week MIT
-
MCP server for the bmdpat.com prompt library — free, tested copy-paste prompts for local LLMs and AI coding agents, inside your Claude/Cursor session. Every prompt links to the post explaining when to
Version check not run yet
Model and agent tools npm Local 4 a week MIT
CM
calyx-mcp io.github.arnab-afk/calyx-mcp-connector 1.5 · 30/100 Connect stdio-only coding agents to hosted Calyx MCP.
Version check not run yet
Coding agent tools npm Local 4 a week MIT
CL
clankerchat io.github.simozampa/clankerchat 1.5 · 30/100 comms for your coding agents
Version check not run yet
Coding agent tools npm Local 4 a week MIT
-
Local-first repository context optimizer for coding agents
Version check not run yet
Coding agent tools npm Local 4 a week MIT
CM
codicil-mcp io.github.pranit-sh/codicil-mcp 1.5 · 30/100 Expose Codicil project knowledge to AI coding agents over the Model Context Protocol.
Version check not run yet
Coding agent tools npm Local 4 a week MIT
-
Local vector retrieval MCP server for coding agents: semantic code search, session-transcript memory, and a live token-savings counter. 100% on-device.
Version check not run yet
Search and retrieval (RAG) npm Local 4 a week MIT
-
Dual-Layer Codebase Knowledge Graph & Episodic Memory Engine for AI Coding Agents
Version check not run yet
Coding agent tools npm Local 4 a week MIT
-
Local developer handoff tool that exports selected Figma design information such as colors, text, spacing, variables, and prototype notes for frontend implementation workflows. Also includes an option
Version check not run yet
Docs and knowledge bases npm Local 4 a week MIT
DA
dravix-agent io.github.rotem1230/dravix-mcp-aegis-v2 1.5 · 30/100 Real-time code-quality gate for AI coding agents (Claude Code) — multi-engine SAST + LLM critic + cross-file taint, OSS, self-hosted.
Version check not run yet
Coding agent tools npm Local 4 a week MIT
DR
drumlin io.github.intelliger-ai/drumlin 1.5 · 30/100 Local-first UX intelligence for AI-assisted development. Reads a Next.js app as a graph of screens, states and actions, and reports the UX problems a type checker cannot see.
Version check not run yet
Coding agent tools npm Local 4 a week APACHE-2.0
-
Give your AI coding agent eyes for Electron apps: launch, drive, screenshot, and get a high-precision pass/fail verdict so it can verify its own UI in a loop.
Version check not run yet
Coding agent tools npm Local 4 a week MIT