Research Tools · Directory rank #23
Agent Cockpit
Keep research claims tied to the code.
Watch the reel
Agent Cockpit is a local whiteboard and MCP server for plans, runs, and claims. It is made for researchers and developers running multi-agent experiments.
Project documentation says deterministic hashes flag changed evidence links, so cards turn red when the code behind a claim changes. Version 0.1.0 is on npm, published 6 November 2026.
Key facts
- Local whiteboard plus MCP server for plans, runs and claims
- Deterministic hashes flag changed evidence links; cards turn red
- Works with Claude Code, Codex, Cursor and Gemini CLI via MCP
- Version 0.1.0, published 6 Nov 2026; needs Node.js 22.13+ and Git; local only
PikyAI has not tested it live. This page is a project overview from the official README and npm page.
PikyAI reel on YouTube. Watch on Instagram
How to use it
- Node.js 22.13 ya us se naya aur Git install karo.
- Agent Cockpit local chalao aur apne agent (Claude Code, Codex, Cursor ya Gemini CLI) ko MCP se connect karo.
- Plans, runs aur claims ek board par record karo; hash checks batayenge ke konsa evidence link badal gaya hai.
Pros and cons
Pros
- Research claims code se juray rehte hain, stale evidence pakra jata hai.
- Checks mein zero extra tokens lagte hain.
- Local only, apna data apne paas rehta hai.
Cons
- Version 0.1.0 experimental hai.
- Node.js 22.13+ aur Git chahiye.
- Sirf local chalti hai, hosted option nahi.
- PikyAI ne live test nahi kiya.
Alternative
- Manual claim tracking — Plans aur claims ko documents ya spreadsheets mein haath se track karna bhi mumkin hai, lekin code change par stale links khud pakarne padte hain.
How this rank works
Directory rank is set by bid, not by editorial opinion. New listings start at $5, a higher bid outranks a lower one, and taking #1 costs at least $5 more than the current top bid. Paid ranks always carry a Sponsored label, and payments are final and non-refundable. This listing sits at the $5 floor as a PikyAI house listing, not a paid rank.