Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add wunki/dotfiles --skill pr-descriptiongit clone --depth 1 https://github.com/wunki/dotfilesWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/wunki/dotfiles/pr-description)<a href="https://agentmods.dev/skills/wunki/dotfiles/pr-description"><img src="https://agentmods.dev/badge/skills/wunki/dotfiles/pr-description/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/wunki/dotfiles/pr-description"><img src="https://agentmods.dev/badge/skills/wunki/dotfiles/pr-description.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00073 | $0.02081 |
| Opus 5.5 | $0.00029 | $0.00832 |
| Sonnet 5.5 | $0.00015 | $0.00416 |
| Haiku 4.5 | $0.00007 | $0.00208 |
Grade A, and why
pr-description scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 1mo ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 256 lines — stays where its author put it; the contents beside it link to each section on GitHub.
PR Description — Context-First Reviewer Narrative
Write PR descriptions that are precise, reviewer-friendly, and narrative: context first, grouped changes second, rationale and notes last. The goal is to make a reviewer understand why this exists, what changed, where to focus, and what risks or edges matter without reading every file first.
Good PR descriptions have a clear thesis, enough background to make the diff obvious, concrete examples where shape or behavior changed, explicit tradeoffs, and direct risk/edge callouts. They respect the reviewer's time by separating narrative from mechanics and leaving routine test status to CI/CD.
This style is calibrated to the user's preferred PR shape. Open directly with the context paragraph—do not add a ## Context heading:
<Context paragraph explaining why the PR exists.>
## What changed
### Focused subsystem / behavior
## Why
## Notes
Use this structure as a default, but shrink or expand it to fit the PR.
Core Style Principles
- Start directly with context, not a heading or implementation. The opening paragraph should explain the situation, constraint, or bug that made the change necessary. Never prefix it with a
Contextheading. - Tell a story. Move from problem → approach → important details → reason/tradeoffs → proof.
- Group by concept, not file. Use
###subsections for meaningful units of change, such as migration, storage model, API behavior, UI state, tests, or rollout. - Be concrete. Include names of important commands, paths, config keys, routes, state keys, data shapes, or user-visible behavior when they help review.
- Show before/after when it clarifies. Use small code/text blocks for schema, filesystem layout, API shape, CLI output, or behavior changes.
- Make ownership and boundaries explicit. Call out what is included, excluded, global vs scoped, backend vs frontend, migration vs steady state, etc.
- Let CI/CD own routine validation. Do not include a default validation/testing section. Mention manual verification only when it is review-relevant and not covered by CI.
- Surface risk and reviewer focus. Put rollback, compatibility, migration, known gaps, manual checks, and contract changes in
## Notesor## Risks. - Avoid generic checklist filler. Prefer compact prose and purposeful bullets over templates like "Summary / Testing" when more context is needed.
- Keep it reviewable. Long PRs can be detailed; small PRs should be concise.
What ships with it
1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 1mo ago First seen · 256 lines · 73 tokens per session scan A be2364d08b74
pr-description is a skill published in the GitHub repository wunki/dotfiles (47 stars, last pushed 13d ago), licensed MIT. It adds 73 tokens to every session and 2,081 once invoked, about $0.0003 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
merge-ready
Drive an existing pull request to a mergeable state: get CI green, resolve merge conflicts with the base branch, address and resolve review comments, trigger required bot reviews/approvals (e.g. commenting '@claude review'), link associated issues, and clean up the PR title and description. Ends with a readiness…
sync-release
Owns release-pipeline conventions: sr.yaml schema, the sr CLI verbs (init/plan/prepare/ release) and urmzd/sr action, typed publishers (cargo/npm/pypi/docker/go/custom), monorepo workspace discovery, versionfiles/stagefiles mapping, and post-release patterns. Language-specific build targets live in scaffold-rust…
setup-ci
Defines GitHub Actions workflow structure: ci.yml + release.yml naming, concurrency groups, bot-commit skip, workflowcall reuse, App-token checkout, and optional fsrc/teasr steps. Language-specific pipelines live in scaffold-rust, scaffold-go, scaffold-python, scaffold-node, scaffold-terraform. Use when setting up…
pr
Open a pull request via gh pr create with an auto-generated title and summary derived from the commits on the branch, grouped by conventional-commit type. Use when the user says "create a pull request", "open a PR", "gh pr create", "raise a PR", or "make a PR for this branch". Assumes commits already exist and only…
release-audit
Audit GitHub releases, git tags, and release assets for one repo or every repo in a directory. Flags orphaned tags, missing binaries/checksums, draft releases, floating tag drift, pre-release remnants, and (when sr.yaml is present) sr config issues. Read-only: never deletes or modifies tags/releases. Use when checking…
repo-status
Scan a folder of git repos and report each repo's local dirty/clean state, ahead/behind divergence from upstream, recent activity, and latest CI run status. Read-only: never commits, pushes, or edits. Defaults to the folder argument or your repo root. Use when checking "what's live", "which repos are dirty", or "repo…