GitHub
Connect AI assistants to GitHub - manage repos, issues, PRs, and workflows through natural language.
Descripción del creador (original en inglés)
Map parallel-agent git worktrees to owners and plan safe cleanup from supplied facts.
Estos tres nunca se combinan en una sola puntuación: las opiniones, las mediciones y la popularidad son cosas distintas.
Todavía no escribimos un resumen sencillo para este. Abajo están los datos que leímos del registro.
Elige la app de IA que usas y copia la configuración.
Necesitas
¿Qué app usas?
Elige tu app arriba. La recordamos, así que la próxima vez el botón dirá “Instalar en …”.
Abre Claude Desktop → Settings → Developer → Edit Config y pega esto en claude_desktop_config.json
{
"mcpServers": {
"treeman": {
"command": "npx",
"args": [
"-y",
"@reneturcios/treeman",
"mcp"
]
}
}
}Ejecuta esto en tu terminal
claude mcp add treeman -- npx -y @reneturcios/treeman mcpPégalo en .cursor/mcp.json de tu proyecto, o en ~/.cursor/mcp.json
{
"mcpServers": {
"treeman": {
"command": "npx",
"args": [
"-y",
"@reneturcios/treeman",
"mcp"
]
}
}
}Tu navegador te pide permiso para abrir la app; luego la app muestra la configuración y te pide confirmar. Los valores de las claves quedan vacíos para que los completes tú.
Pégalo en .vscode/mcp.json de tu proyecto
{
"servers": {
"treeman": {
"type": "stdio",
"command": "npx",
"args": [
"-y",
"@reneturcios/treeman",
"mcp"
]
}
}
}Tu navegador te pide permiso para abrir la app; luego la app muestra la configuración y te pide confirmar. Los valores de las claves quedan vacíos para que los completes tú.
Este se ejecuta en tu computadora, y ChatGPT solo puede conectarse a servidores en internet. Elige otra app.
Tu elección se guarda en una cookie en este dispositivo; no necesitas cuenta.
Esta configuración se genera a partir de la ficha del creador en el registro. Si el README dice otra cosa, sigue el README.
Se ejecuta en tu computadora; la lista de herramientas aparece en tu app de IA después de instalarlo.
Revisamos información que cualquiera puede ver; no hemos leído el código. Las revisiones que no aplican no se cuentan. · Revisado el 6 nov 2026
El nombre en el registro se verificó con la cuenta de GitHub CloudCorpRecords, dueña del repo
Todavía no hemos leído este repo en GitHub
Todavía no hemos leído este repo en GitHub
Este se ejecuta en tu computadora; no hay endpoint en internet
Se ejecuta en tu computadora y usa ajustes locales en su lugar
Se ejecuta en tu computadora; solo podemos leer listas de herramientas de endpoints en internet
Comparado con todos los servidores activos de esta categoría en el registro. Las estrellas solo cuentan cuando quien lo publica es dueño del repo.
Superó 1 de 1 revisiones básicas aplicables. No hemos leído el código. Antes de agregar una clave o un token, revisa qué acceso pide el servidor y usa una clave que tenga solo los permisos que necesitas.
El registro no indica ninguna clave obligatoria, aunque algunos servidores te piden iniciar sesión la primera vez que los usas.
Se ejecuta en tu computadora. Tu app de IA lo inicia cuando hace falta, y los datos no pasan por los servidores del creador (a menos que el propio servidor llame a un servicio en línea).
El repo de GitHub cambió por última vez –, y la última versión (0.1.0) llegó al registro hace 3 días.
Las apps compatibles con MCP, como Claude Desktop, Claude Code, Cursor y VS Code. También necesitas Node.js, Docker en tu computadora.
En la misma categoría: GitHub, Desktop Commander, Cloudflare MCP
Servidores de la misma categoría hechos por otros creadores, ordenados por estrellas.
Cómo leer una tarjeta:★ Estrellas de GitHubiTres pilares: reseñas · revisiones · usoiActualizado = fecha en que se publicó la última versión en el registroi
Connect AI assistants to GitHub - manage repos, issues, PRs, and workflows through natural language.
MCP server for terminal commands, file operations, and process management
Cloudflare MCP servers
Copiado tal cual del repo del creador. PopNx MCP no lo ha revisado ni lo respalda. Se quitaron HTML, imágenes y scripts. Los enlaces llevan fuera de este sitio.
When several agents work in the same repository at once, each one usually gets its own git worktree, and after a while nobody remembers which tree belongs to whom, which ones still hold uncommitted work, and which are safe to delete. treeman answers those questions from facts you already have. You supply, for each worktree, its path, the branch checked out in it, who owns it if anyone does, whether it has uncommitted changes, how many days have passed since it last changed, and whether it is locked — plus the inactivity threshold beyond which a tree counts as stale, and optionally the path of the primary tree. treeman hands back the plan: an ownership map, a status and cleanup action for every tree, and the tallies. It never runs git and never touches the filesystem; it plans from the supplied facts only.
Each tree is judged by a fixed set of checks, first match wins. A tree whose path is the supplied main_path is main and is always kept. A locked tree is locked and kept. A dirty tree is blocked and kept — uncommitted changes are never planned away. A tree that is both past the stale threshold and unowned is orphaned, and it is the only kind planned for removal. An owned tree past the threshold is stale and marked for review; a recent tree with no recorded owner is unowned and also reviewed; an owned, recently active tree is active and kept. The staleness comparison is strictly greater-than: a tree whose last activity is exactly stale_days ago is not stale.
Ownership has one pinned judgement call: an owner that is absent, empty, or only whitespace means the tree is unowned, and its owner is reported as null. The owners map in the result lists every other owner with the paths it owns in the order you supplied them, with the owner keys sorted ascending; unowned trees never appear in it. The counts report the total, one tally per status, and removable — how many trees the plan marks remove.
Malformed input — a missing or empty worktree list, a missing, empty, or non-string path or branch, a non-string owner, a missing or non-boolean dirty, a non-boolean locked, a negative or fractional day count, a bad stale_days, or an empty main_path — is a validation error naming the offending path, with no partial result. Two worktrees with the same path are an error naming the duplicated path. Unknown extra fields are ignored.
npx @reneturcios/treeman plan --input args.json
npx @reneturcios/treeman --version
args.json contains the full arguments object:
{
"worktrees": [ { "path": "/trees/main", "branch": "main", "owner": "alice", "dirty": false, "last_activity_days": 0 }, { "path": "/trees/old-experiment", "branch": "exp/spike", "dirty": false, "last_activity_days": 74 }, { "path": "/trees/agent-2", "branch": "feat/search", "owner": "agent-2", "dirty": true, "last_activity_days": 1 } ], "stale_days": 30, "main_path": "/trees/main" }
The plan is printed as JSON:
{
"worktrees": [ { "path": "/trees/main", "branch": "main", "owner": "alice", "status": "main", "cleanup": "keep", "reason": "primary worktree" }, { "path": "/trees/old-experiment", "branch": "exp/spike", "owner": null, "status": "orphaned", "cleanup": "remove", "reason": "no owner and inactive beyond the stale threshold" }, { "path": "/trees/agent-2", "branch": "feat/search", "owner": "agent-2", "status": "blocked", "cleanup": "keep", "reason": "uncommitted changes present" } ], "owners": { "agent-2": ["/trees/agent-2"], "alice": ["/trees/main"] }, "counts": { "total": 3, "main": 1, "locked": 0, "blocked": 1, "orphaned": 1, "stale": 0, "unowned": 0, "active": 0, "removable": 1 } }
Exit codes: 0 on success, 1 on a runtime error (including an unreadable or invalid input file, or invalid input), 2 on a usage error.
npx -y @reneturcios/treeman mcp
MCP client config:
{
"mcpServers": { "treeman": { "command": "npx", "args": ["-y", "@reneturcios/treeman", "mcp"] } } }
plan_worktrees — Plan parallel-agent git worktrees from caller-supplied worktree facts: the ownership map, each tree's status, and which trees are safe to remove.worktrees (array, required, at least 1 item) — the worktrees to plan, in the caller's order; each item:path (string, required, non-empty) — worktree pathbranch (string, required, non-empty) — branch checked out in the worktreeowner (string, optional) — agent or person who owns the worktree; absent, empty, or whitespace-only means unowneddirty (boolean, required) — whether the worktree has uncommitted changeslast_activity_days (integer, required, at least 0) — days since the worktree last changedlocked (boolean, optional) — whether the worktree is locked; absent means falsestale_days (integer, required, at least 0) — inactivity threshold in days; beyond it a worktree is stale (exactly at it is not)main_path (string, optional) — path of the primary worktree, which is never cleaned upReturns {worktrees, owners, counts}, where worktrees holds one {path, branch, owner, status, cleanup, reason} per input worktree in input order (owner is null when unowned), owners maps each distinct owner to its paths, and counts is {total, main, locked, blocked, orphaned, stale, unowned, active, removable}.
docker run -i ghcr.io/cloudcorprecords/treeman
docker run --rm -v "$PWD:/w" ghcr.io/cloudcorprecords/treeman plan --input /w/args.json
npm install
npm run build # tsc -> dist/ npm test # tsx --test test/*.test.ts (pure local, no network) npm run dev # run the CLI from source
MIT — Copyright (c) 2026 Rene Turcios