writing-beats
An agent skill by mattpocock, from mattpocock/skills. Tags: analysis, content-strategy, copywriting, documentation, writing.
What it does
Writing, exploit; assemble raw material into a journey of beats, grounding each term before a beat leans on it.
Install
With the skills CLI, which installs into Claude Code, Codex, Cursor and other agents:
npx skills add mattpocock/skills --skill writing-beatsOr 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/mattpocock/skills
cp -r skills/skills/in-progress/writing-beats ~/.claude/skills/writing-beatsSafety box score
Grade B, low risk: 11 of 100 risk points, rated 2026-10-16. 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 guided writing workflow that reads a markdown source file and repeatedly writes one beat at a time into an article file. It does not ship code execution, network access, or privilege changes; the main risk is whatever file path the user chooses for the article, since the agent will read and overwrite that file during the drafting loop.
| Category | Level | Why |
|---|---|---|
| Code & shell execution | 0 (none) | No shell commands, code evaluation, or bundled scripts are present; the skill only describes a writing workflow in `SKILL.md`. |
| Network access | 0 (none) | There are no download, upload, API, or remote-host interactions in the shipped files. |
| Filesystem reach | 2 (moderate) | The skill repeatedly reads and writes an "article file" at a user-chosen path, with no restriction that it stay inside the repo: "ask once and remember the path" and "write only that beat to the article file." |
| Secrets & credentials | 0 (none) | It does not mention tokens, passwords, keychains, `.env` files, or other secret stores. |
| Privilege & persistence | 0 (none) | No sudo, global installs, services, scheduled tasks, or shell/OS configuration changes are requested. |
| Instruction hijack surface | 0 (none) | The files do not instruct the agent to ignore higher-priority rules, follow remote instructions, hide content, or otherwise override its normal controls. |
| Irreversible actions | 1 (low) | The workflow is interactive and pauses for user choices at each beat; it does not autonomously publish, send messages, or take external actions beyond editing the article file. |
| Transparency | 0 (none) | All shipped content is plain text (`SKILL.md` and `agents/openai.yaml`); there are no binaries, generated blobs, or unreadable installers. |
Before installing
- Confirm the article save path is a scratch file in a safe directory, not a dotfile or important document.
- Read `SKILL.md` and verify the only writes are to the article file you choose.
- Test it once with a dummy markdown source and a disposable output file first.
Rated 2026-10-16 by an automated review of 2 files from the skill's repository.
Source
- Repository
- mattpocock/skills (all skills from this repository)
- Path
- skills/in-progress/writing-beats/SKILL.md
- Branch
- main
- Updated
- 2026-10-05
Related skills
- writing-shape — Writing, exploit: shape raw material into an article, paragraph by paragraph.
- bro — Restate the last message in plain human language, with no jargon.
- teach — Explain a body of work plainly so a person actually understands it. Runs the `how` and `why` skills and weaves what they find into one clear explanation.
- writing-fragments — Writing, explore: mine raw fragments, no structure yet.
- audit-prep-assistant — Prepares codebases for security review using Trail of Bits' checklist.
- automate-me — Use for "automate me", "create/update/refresh my -mode skill", "turn/capture my preferences or working style into a skill".