PAPI + Claude Code

PAPI runs inside Claude Code as an MCP server. The plan, the decisions, and the direction live in the same session that writes the code, and PAPI flags the moment new work slides off them.

Connect PAPI See what PAPI does

Free on three projects · no card · 14-day refund on Pro

How PAPI fits Claude Code

No new app to open: PAPI runs inside the AI tool you already build in. Type plan, orient, or review as chat messages. PAPI's tools show up in Claude Code like any other MCP server, no tab switch, no separate board to keep current.

Decisions on the record: every call your project makes is kept, with the why, and checked against new work. Start a fresh session tomorrow and Claude Code reads them back before it touches related work. No re-explaining the architecture you settled last week.

Catches drift: the moment new work slides off the plan, PAPI flags it. Claude Code is fast enough to build the wrong thing convincingly. Every build lands against a handoff with scope and acceptance checks, so done means checked, not claimed.

Connect PAPI to Claude Code

Create your account. Sign in at getpapi.ai with GitHub or email and name your first project.

Paste this into Claude Code. Claude Code adds PAPI itself and opens a browser tab so you can sign in. No terminal, no token to copy.

Paste into Claude Code chat

Add PAPI as an MCP server for this project. Run this from the project root:

  claude mcp add --transport http papi https://mcp.getpapi.ai/mcp

The server is added but NOT yet authenticated — only you can finish this, in a browser. After the command succeeds, tell me exactly this:
"PAPI is added. Sign in to connect it: run  claude mcp login papi  (Claude Code v2.1.186+) and approve in the browser — or run /mcp, choose papi, pick Authenticate. Sign in with the same account you used on getpapi.ai."

Run setup. Ask Claude Code to run setup. PAPI writes your product brief, then plan starts your first cycle.

Questions

Claude Code has memory now. Why add PAPI?: Memory recalls what happened. It doesn't hold a plan, sequence tasks, gate builds, or tell you when the work slid off course. A memory of everything you did is not a plan for what's next.
My CLAUDE.md already does this.: Files describe. They don't decide. Nothing checks the finished work against them, and they never reorder the backlog when reality changes.
Does PAPI read my code?: Your code stays on your machine. PAPI adds a planning layer to whatever you're already building, it doesn't read your source.
Can I use the same project from other tools?: Yes. PAPI is one remote MCP server, so the same project connects from Claude Code, Cursor, Codex, or any MCP tool, and every session picks up the same plan.