lpm skill
Three lpm skills: the agent writes project config, runs services and repo copies, and keeps shared work notes.
- 150 ★ on the repository
- Last commit
- MIT
Extended access
Why this level
- The `lpm-cli` skill starts and stops services and creates copies and Git worktrees; commands between computers need approval in the app.
Install
- Works in
In your terminal, with SkillFoxx CLI
npx skillfoxx add skills/lpmDetects the agents on your machine, shows the access level and pins the version.
Other ways to install
Access and checks
How we reviewExtended accessThe tool runs code or makes network calls and only reads external data.
- The `lpm-cli` skill starts and stops services and creates copies and Git worktrees; commands between computers need approval in the app.
This is third-party code. Review the repository files before installing.
What it does
lpm is a desktop app that runs project services and AI agents in one window. The repository ships three skills for it. `lpm-config` writes and edits project config, services and action buttons. `lpm-cli` starts and stops projects, reads logs, and creates copies and linked Git worktrees. `lpm-memory` saves work notes to `~/.lpm/memory/` so another agent can continue from the same point. The config skill changes files only through `lpm config apply`: the command validates the candidate first, then the app replaces the file atomically.
Who it is for. Developers running Claude Code and Codex in parallel on several project copies who want the agents themselves to drive the app.
Good fit when
- A project has several services and the launch config is easier to hand to an agent.
- Each task needs its own repository copy so agents do not collide.
- One agent finishes a session and another must pick up the notes and continue.
- An agent has to read service logs or wait until a port opens.
Not a fit when
- The lpm app is not installed, and `lpm-cli` does not work without it.
- You need to run without a desktop window, such as on a headless server.
- Plain `docker compose` is enough and a separate project manager is overhead.
Example request
Set up lpm for this project: start the server and worker, and add a button for tests.Limitations
The skills work only with the lpm app: download it from lpm.cx first, and install the `lpm` command from the app's Settings. Several commands (`config get`, `config apply`, `skills`, project control) need the app running. On macOS config is edited only through `lpm config apply`, while the README marks Windows and Linux as beta. The `lpm cp` and `lpm exec` commands between computers wait for user approval in the app and are off by default.
How to disable. Delete the `lpm-config`, `lpm-cli` and `lpm-memory` folders from the agent's skills directory. The lpm app and notes in `~/.lpm/memory/` stay and are removed separately.
FAQ
How do I install the skills?
Run `npx skills add gug007/lpm` and tell the agent "set up lpm for this project". The lpm app itself is downloaded separately from lpm.cx.
How do the skills differ?
`lpm-config` edits the YAML config, `lpm-cli` controls running projects and copies, and `lpm-memory` keeps work notes for handoff between agents.
Where does `lpm-memory` write notes?
To `~/.lpm/memory/<project>/<session>.md`; `$LPM_MEMORY_DIR` can set the path. A project copy shares notes with its original.
Can I edit the config by hand?
The skill does not: it reads a layer with `lpm config get` and writes through `lpm config apply` with a revision check, so the app never receives a broken file.
Do the skills work in Codex?
Yes, `lpm-cli` ships `agents/openai.yaml` and installs with the same `skills add` command.
Discussion
No questions or reviews yet.
Related
SkillsAPM
APM (Agent Package Manager)
A package manager for AI agent context: skills, prompts and MCP servers install with one command across supported agents
The agent-lsp MCP server orchestrates language servers into agent workflows: callers, safe edits, renames and diagnostics.
Skills and a CLI with a product graph: roles, states and failures checked before code, the live product after the build
A package manager for AI agents that installs rules, skills and prompts and converts them for Cursor, Claude Code and others.
A Composer plugin that ships AI skills as dependencies, pulls them from vendor and Git and syncs to a project folder.
An extension that gives the Pi agent LSP, linters, formatters and structural search on every code edit, plus an experimental MCP.
README in short
The skills install with one command, after which the agent configures and drives the project through the `lpm` command.
README badge
Original SKILL.md
--- name: lpm-memory description: "Shared project memory for AI coding agents: save or recall work-session logs in `~/.lpm/memory/<project>/<session>.md` so another agent CLI (Claude Code, Codex, Gemini) or a future session can continue the work by session name." ---