How do I install RSI Boot?

RSI Boot is a local plugin. Install it using PyPI package: rsi-boot and add the generated configuration snippet to your AI app's MCP config file. Then restart your AI app.

Is RSI Boot safe to use?

RSI Boot was flagged by MCP Marketplace's security scan, scoring 4.6/10 (high risk). It has 5 high or critical findings to review. Review the security report on this page carefully before installing it.

What credentials does RSI Boot need?

RSI Boot requires the following credentials or environment variables: RSI_PROJECT_ROOT. You can find setup instructions on the server detail page.

What AI apps work with RSI Boot?

RSI Boot uses the Model Context Protocol (MCP) and works with any MCP-compatible AI app, including Claude, ChatGPT / Codex, Gemini, Copilot, Cursor, and more.

Back to Browse

RSI Boot MCP Server

by User

Developer ToolsUse Caution4.6LocalNew

Free

Repo-bound memory and a self-improving harness for AI coding agents — zero API keys.

About

RSI Boot is a local MCP server that gives AI coding agents persistent, repo-bound memory. Conventions, prohibitions, fix lessons and harness tunables all live in the project's own `.rsi/` directory, so memory moves with the repository — across machines and across hosts (Cursor, Claude Code, Codex, Qoder) with nothing to relearn.

Unlike plain memory plugins, RSI Boot also improves the harness itself: negative feedback goes into failure buckets, improvement proposals must pass static gates and human approval, are observed for 7 days, and roll back automatically if acceptance drops. The main path calls no model and needs no API key; every write is gated by approval workflows, prohibition lists and conflict adjudication.

Built for developers and teams who use AI coding agents daily and want per-project memory that is portable, auditable, and gets better the more they use it.

Security Report

4.6

Use Caution4.6High Risk

RSI Boot is a repo-bound local memory MCP server whose README and visible code show no exfiltration, backdoor, or hardcoded credentials; main-path operation is zero-API-key and permissions (filesystem under .rsi/, optional network for model features) match its developer-tool purpose. Only minor concerns remain: the bootstrap scanner reads conversation context and project files on user consent, and the visible code is truncated so the full network and subprocess usage could not be verified. Supply chain analysis found 7 known vulnerabilities in dependencies (0 critical, 5 high severity). Package verification found 1 issue.

5 files analyzed · 11 issues found

Security scores are indicators to help you make informed decisions, not guarantees. Always review permissions before connecting any MCP server.

Permissions Required

This plugin requests these system permissions. Most are normal for its category.

File System Read

Reads files on your machine. Normal for tools that analyze or process local data.

File System Write

Writes or modifies files on your machine. Check that this is expected for the tool.

env_vars

Check that this permission is expected for this type of plugin.

HTTP Network Access

Connects to external APIs or services over the internet.

process_spawn

Check that this permission is expected for this type of plugin.

database

Check that this permission is expected for this type of plugin.

What You'll Need

Set these up before or after installing:

Workspace to bind memory to. Optional — falls back to MCP roots and the server working directory when unset. For Cursor, set it only in project-level .cursor/mcp.json; ${workspaceFolder} does not expand in user-level configs.Optional

Environment variable: RSI_PROJECT_ROOT

How to Install

Add this to your MCP configuration file:

{
  "mcpServers": {
    "rsi-boot": {
      "env": {
        "RSI_PROJECT_ROOT": "your-rsi-project-root-here"
      },
      "args": [
        "rsi-boot"
      ],
      "command": "uvx"
    }
  }
}

Getting Started

Once installed, try these example prompts and explore these capabilities:

1"Learn this project" — bootstrap distills repo knowledge into reviewable packs
2"What conventions should I follow in this repo?"
3"Remember: this project uses snake_case everywhere"
4Tool: rsi_recall — Recall prohibitions and conventions before starting a task
5Tool: rsi_learn — Distill project knowledge and record lessons
6Tool: rsi_feedback — Report whether recalled guidance was accepted or rejected
7Tool: rsi_knowledge_review — Approve distilled knowledge before it enters recall