automate-me
An agent skill by cursor, from cursor/plugins. Tags: analysis, automation, documentation, skill, workflow.
What it does
Use for "automate me", "create/update/refresh my -mode skill", "turn/capture my preferences or working style into a skill", or wanting agents to follow how the user works. Drafts or revises a personal -mode skill via create-skill + unslop, optionally pulling fresh evidence from recent transcripts.
Install
With the skills CLI, which installs into Claude Code, Codex, Cursor and other agents:
npx skills add cursor/plugins --skill automate-meOr copy the skill folder into Claude Code's skills directory by hand (~/.claude/skills for every project, or .claude/skills inside one):
git clone --depth 1 https://github.com/cursor/plugins
cp -r plugins/pstack/skills/automate-me ~/.claude/skills/automate-meSafety box score
Grade C, moderate risk: 26 of 100 risk points, rated 2026-10-19. The grade rates what the skill and its scripts can reach on the machine of whoever installs it, across eight categories from shell execution to secrets access. How the score works.
This skill is a meta-workflow for drafting or revising a personal `-mode` skill by reading existing skills and recent workspace transcripts, then iterating with the user and landing the result. The main risks are that it can read personal history from `agent-transcripts/` and home-directory skill locations, and it ends by committing and opening a PR, which publishes the generated skill remotely.
| Category | Level | Why |
|---|---|---|
| Code & shell execution | 0 (none) | The file does not ship scripts or shell commands; it only orchestrates other skills and agent steps like `create-skill` and `unslop`. |
| Network access | 2 (moderate) | It explicitly ends with remote publication via "Commit and open a PR," which sends the drafted skill to a git host using the user's credentials. |
| Filesystem reach | 2 (moderate) | It reads and may edit skill files under `.cursor/skills/**`, `~/.cursor/skills/*-mode/SKILL.md`, and the workspace `agent-transcripts/` directory, including recursive home-directory scans. |
| Secrets & credentials | 1 (low) | It mines recent transcripts as source material, so it may encounter pasted credentials or other sensitive text, though it does not directly request keychains, `.env` files, or token stores. |
| Privilege & persistence | 0 (none) | There is no sudo, service registration, global install, or OS/shell-setting change in the skill text. |
| Instruction hijack surface | 0 (none) | It contains no instructions to ignore safeguards, hide behavior, or follow fetched remote instructions; the only system-prompt reference is a path lookup for transcripts. |
| Irreversible actions | 2 (moderate) | It directs the agent to survey history, ask questions, draft the skill, iterate, work in a worktree, and then "Commit and open a PR," which is a meaningful autonomous publish workflow. |
| Transparency | 0 (none) | The skill is plain text only; there are no binaries, minified blobs, base64 payloads, or unreadable installers. |
Before installing
- Confirm you're okay with the skill reading `agent-transcripts/` in the active workspace.
- Decide whether it may inspect or edit `~/.cursor/skills/*-mode/SKILL.md` on your machine.
- Verify you want the agent to create commits and open a PR on your behalf.
- If you have sensitive material in recent chats, review or redact those transcripts before using this skill.
Rated 2026-10-19 by an automated review of 1 file from the skill's repository.
Source
- Repository
- cursor/plugins (all skills from this repository)
- Path
- pstack/skills/automate-me/SKILL.md
- Branch
- main
- Collection
- cursor
- Updated
- 2026-09-23
Related skills
- workflow-from-chats — Extract durable working preferences from recent Cursor chats and convert them into skills, rules, or workflow docs.
- continual-learning — Orchestrate continual learning by delegating transcript mining and AGENTS.md updates to `agents-memory-updater`.
- principle-build-the-lever — Apply to any non-trivial work, not just bulk work: edits, migrations, analyses, checks.
- recall — Reconstruct your recent working context from your own chat history, live state, and the shared record (user reports, prior fixes, incidents).
- show-me-your-work — Keep a reviewable decision trail for long-running or unattended work: a TSV log with one row per decision (what, why, evidence, result).
- swarm — Fan out N parallel workers, drain them, and return one report. Use for /swarm, 'swarm this', or parallel coverage, races, gauntlets, and exploration.