Flanner
Version-controls the plan files Claude Code and Codex write, and flags stale ones
- Pricing
- Freemium
- Platforms
- CLI, self-hosted
- Runs
- Locally
- Version
- 0.16.0
- Listing updated
- 1 day ago
About Flanner
Flanner is a freemium MCP server for developer tools. It runs locally on your machine and is started by your AI client.
Flanner treats the markdown plans that coding agents produce as documents with a history. Each edit becomes a new numbered version, you can diff any two, and a status backed by evidence (dead references, commits since it was written, age) tells you when a plan has gone stale. Plans live in a .plans/ directory that is kept out of git until you decide otherwise.
It works through three surfaces on one local SQLite catalogue: a CLI, the MCP server your agent drives, and a local web UI. Running flanner init registers it with Claude Code and Codex and adds a guard hook plus managed CLAUDE.md and AGENTS.md blocks so agents save plans through it. Add-ons cover searchable project memory that refuses secrets, a scan of which skills an agent actually loads, and Curb, which maps which credentials and channels an agent can reach. Curb deliberately has no MCP tools.
The local tool needs no account. A paid Mesh tier syncs plans, memory and skills between your own devices or a team device to device, with the hosted side handling accounts and access rather than content.
Install
Run in a terminal
claude mcp add flanner -- uvx flannerAdd to claude_desktop_config.json
{
"mcpServers": {
"flanner": {
"command": "uvx",
"args": [
"flanner"
]
}
}
}Add to ~/.cursor/mcp.json
{
"mcpServers": {
"flanner": {
"command": "uvx",
"args": [
"flanner"
]
}
}
}Add to .vscode/mcp.json
{
"servers": {
"flanner": {
"type": "stdio",
"command": "uvx",
"args": [
"flanner"
]
}
}
}Add to ~/.codeium/windsurf/mcp_config.json
{
"mcpServers": {
"flanner": {
"command": "uvx",
"args": [
"flanner"
]
}
}
}Add to your client's MCP settings (most use this shape)
{
"mcpServers": {
"flanner": {
"command": "uvx",
"args": [
"flanner"
]
}
}
}What it does
- Versioned plans
- Every revision is kept with full history and can be diffed.
- Staleness status
- Reports dead references and commits since a plan was written.
- Git protection
- Keeps plan files out of commits automatically.
- Agent integration
- Init adds a guard hook and CLAUDE.md and AGENTS.md guidance.
- Project memory
- Saves decisions and constraints as searchable markdown and refuses secrets.
- Curb reports
- Shows which credentials and channels an agent can reach, as counts rather than names.
- Mesh sync
- Optional paid device-to-device sync for individuals and teams.
Good for
- Keeping design docs written by agents from piling up and rotting
- Seeing what a plan said before the agent changed its mind
- Sharing plans and memory across a team without a central store
- Auditing what a coding agent can read on your machine
Pricing
Freemium, from $9 a month
Local tool free (MIT); Personal Mesh $9/month with 30-day trial; Team and Organization on request
Prices change. Check flanner.io before you buy.
Flanner FAQ
Is it free?
The CLI, MCP server and local web UI are free and open source under MIT; Mesh sync is a paid add-on.
Do I need an account?
No, not for the local tool. An account matters only for Mesh.
How do I install it?
uv tool install flanner, then flanner init in your project.
What does Personal Mesh cost?
$9 a month for one person on up to 5 devices, with a 30-day trial.
Which assistants does it support?
Claude Code and Codex out of the box, plus any MCP client.
Is Flanner free?
Flanner has a free plan, with paid plans for more usage or features. Paid plans start at $9 a month.
How do I add Flanner to Claude Code?
Run this command in your terminal: claude mcp add flanner -- uvx flanner
What are the best alternatives to Flanner?
The closest alternatives to Flanner are REA, Codebase Memory, GitHub, Serena MCP: the IDE for your agent and XERJ.
Alternatives to Flanner

REA
MCP serverCLI and MCP server that lets coding agents reverse engineer apps and binaries
Runs locally- Open source
- 464K GitHub stars
- Developer tools

Codebase Memory
MCP serverLocal MCP server that builds a code knowledge graph for AI coding agents
Runs locally- Open source
- 48.7K GitHub stars
- Knowledge and memory

GitHub
MCP serverOfficial MCP server connecting AI assistants to GitHub repos, issues and workflows
Not checked yet- Remote
- Open source
- 34.5K GitHub stars
- Developer tools

Serena MCP: the IDE for your agent
MCP serverMCP toolkit giving coding agents symbol-level code retrieval and editing
Runs locally- Open source
- 31.5K GitHub stars
- Developer tools

XERJ
MCP serverLocal search engine for AI agents with cited answers over your code and docs
Runs locally- Open source
- 19.2K GitHub stars
- Knowledge and memory

Ouroboros
MCP serverSpec-first layer that pins requirements and verifies results of AI coding agent work
Runs locally- Open source
- 6.4K GitHub stars
- Developer tools