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 tellmefrankie/news-engine --skill commit-roastgit clone --depth 1 https://github.com/tellmefrankie/news-engineWrote 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/tellmefrankie/news-engine/commit-roast)<a href="https://agentmods.dev/skills/tellmefrankie/news-engine/commit-roast"><img src="https://agentmods.dev/badge/skills/tellmefrankie/news-engine/commit-roast.svg" alt="Measured on agentmods" 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 | $0.00000 | $0.00818 |
| Opus 5 | $0.00000 | $0.00409 |
| Sonnet 5 | $0.00000 | $0.00164 |
| Haiku 4.5 | $0.00000 | $0.00082 |
Grade A, and why
commit-roast 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 6d 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 — 95 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Commit Roast
A savage but accurate code review of your git commit history. Identifies real problems through the lens of a brutally honest senior engineer.
What it does
Reads your recent commits and delivers a roast that is 30% comedy, 70% real feedback:
- Calls out vague messages ("fix", "update", "wip", "stuff")
- Flags commit size issues (300-file commits, or 47 commits to change one line)
- Spots patterns: late-night commits, "fix previous fix" chains, whitespace-only changes
- Identifies the commit where you clearly gave up (usually commit 23 of 25 on a Friday)
- Ends with 3 genuine suggestions to improve your commit hygiene
Example Output
Commit Roast — last 30 commits
THE HIGHLIGHTS:
"fix" — May 9, 2:47 AM
A 47-file commit at 3 AM called "fix." Fix what? Fix everything?
Fix your sleep schedule? We may never know.
"wip" (x4) — May 10
Four consecutive WIP commits. You committed "wip" to main.
To the main branch. Four times.
This is not a draft. This is a published novel.
"update utils" — May 11
Updated utils.ts. All 847 lines of it.
One commit. 847 lines.
Reviewers saw this and went home early.
"fix previous fix" — May 12
The ouroboros commit. The fix that ate its own tail.
The commit that proves the previous commit should not have existed.
WHAT YOU DID WELL:
- Commits on May 8 were actually good. Specific, scoped, readable.
"feat(auth): add refresh token rotation with 7-day expiry" — chef's kiss.
- You rebased before the PR. Respect.
GENUINE FEEDBACK:
1. Your commit messages average 4 words. Aim for 8-12. Subject + brief why.
2. Three "wip" commits = one real commit. Squash before pushing.
3. 2 AM commits average 3x more "fix" commits the next morning. Sleep first.
VERDICT: B- (was A- before the 3 AM incident)
Usage
Roast my last 30 commits.
Be honest about what is bad, but make it funny.
End with 3 real actionable suggestions.
Severity: medium (not mean, not toothless)
Severity modes
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.
- 6d ago First seen · 95 lines · 0 tokens per session scan A 1aacdba3497f
commit-roast is a skill published in the GitHub repository tellmefrankie/news-engine (1 stars, last pushed 3mo ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 818 tokens. 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 2026-08-31.
Other skills, from other repositories
lean-review
CocoLean diff-scoped over-engineering audit — scans uncommitted git diff and applies five classification tags (delete/stdlib/native/yagni/shrink) to identify unnecessary surface area before commit.
flag-cleanup
Weekly reusable-session feature-flag cleanup for {{targetrepo}}. Inventories every feature flag, classifies each as fully rolled out, long dead, partial rollout, or active experiment, removes the flag and the dead branch it guards on an isolated branch when the suite is green, and opens a PR — never touching a flag…
github-commenting
How to post clean, rich, deduplicated GitHub PR review comments — suggestion blocks, multi-line anchors, markers, formatting rules. Load before posting or fixing any PR comment.
open-pr-fix
Act on the findings a review left on a PR or MR: take or decline each by severity, edit the code, one commit, reply. Use when handed a PR or MR URL that has already been reviewed. Edits real code.
vanellus
Skill "vanellus" from markmhendrickson/ateles, covering vanellus — pr steward, identity, principals, core job and automated pr review (swarm-panel invocation).
github
GitHub operations via gh CLI: issues, PRs, CI runs, code review, API queries. Use when: (1) checking PR status or CI, (2) creating/commenting on issues, (3) listing/filtering PRs or issues, (4) viewing run logs. NOT for: complex web UI interactions requiring manual browser flows (use browser tooling when available)…