Skills — independently scanned and version-tracked by SaferSkills.
SaferSkills independently audited Skills (Plugin) and scored it 100/100 (green). The audit ran 55 deterministic rules across Security, Supply Chain, Maintenance, Transparency, and Community; it found 0 high-severity and 0 lower-severity findings. The full rule-by-rule trace and per-finding evidence are below. Free, methodology-open.
Findings & checks · 0 flagged
Every scanned point with the score it earned and what moved between them.
First recorded scan — no prior version to compare against.
The primary manifest — the file an agent reads to learn what this artifact does.
Brandt's authored and customized agent skills.
These skills follow the Agent Skills specification, so any skills-compatible agent — Claude Code, Codex CLI, and others — can use them. Everything installs straight from GitHub: no npm, no third-party installer. Validated in CI with skills-ref.
An illustrated walkthrough is published at [brandtam.github.io/skills](https://brandtam.github.io/skills/INSTALL.html) (source: INSTALL.html).
/plugin marketplace add brandtam/skills
/plugin install brandtam-skills@brandtamgit clone https://github.com/brandtam/skills.git
cd skills
./install.sh # all skills -> ~/.agents/skills
./install.sh write-commit # or just the ones you wantTarget a specific agent's skills directory instead:
DEST=~/.codex/skills ./install.sh # Codex CLI
DEST=~/.claude/skills ./install.sh # Claude CodeEach skill is a self-contained folder under skills/. Copy any one into your agent's skills directory:
~/.codex/skills/~/.claude/skills/~/.agents/skills convention) — ~/.agents/skills/cp -R skills/write-commit ~/.codex/skills/| Skill | Description |
|---|---|
| prepare-release | Cut a release from pending changesets — version preview, changelog, archived notes, tag / GitHub Release steps. |
| write-changeset | Write or update a changeset for a PR or branch, scaffolding the workflow if it's missing. |
| write-commit | Generate structured commit messages and optionally commit, honoring templates and requiring approval. |
| uat | Run a user-acceptance-testing session — collect observations, grill for decisions, generate a PRD. |
| resurrect-project | Revive a dormant project — survey the code, interview on today's intent, scaffold a rewrite plan. |
Check every skill against the spec before committing or releasing:
./validate.shIt installs skills-ref into a local .venv (Python, no npm) and validates each folder under skills/. The same check runs in CI on every push and PR.
.claude-plugin/
marketplace.json # marketplace catalog (lists this repo as one plugin)
plugin.json # this repo IS the plugin; skills auto-discovered
skills/
<name>/SKILL.md # one self-contained, spec-compliant folder per skill
install.sh # copy skills into a skills directory (bash, no npm)
validate.sh # validate skills against the Agent Skills spec
.github/workflows/ # CI: runs validation on push/PR
INSTALL.html # illustrated walkthrough~30 seconds. Free. No account. Every finding cites a rule and a line of evidence.