Getting it into your agent
It runs from inside its repository, so the clone comes first — what it calls does not travel with the file alone.
git clone --depth 1 https://github.com/rikdc/ai-skillsnpx agentmods add skills/rikdc/ai-skills/forgejo-advisory-reviewWrote 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/rikdc/ai-skills/forgejo-advisory-review)<a href="https://agentmods.dev/skills/rikdc/ai-skills/forgejo-advisory-review"><img src="https://agentmods.dev/badge/skills/rikdc/ai-skills/forgejo-advisory-review/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/rikdc/ai-skills/forgejo-advisory-review"><img src="https://agentmods.dev/badge/skills/rikdc/ai-skills/forgejo-advisory-review.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 | $0.00086 | $0.02367 |
| Opus 5.5 | $0.00034 | $0.00947 |
| Sonnet 5.5 | $0.00017 | $0.00473 |
| Haiku 4.5 | $0.00009 | $0.00237 |
Grade C, and why
forgejo-advisory-review scanned grade C with 2 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 2d 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.
Recursive force deletehighDestructive command
rm -rf with a variable or a broad path is one typo away from removing the wrong tree.
+ rm -rf ./cache/* Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
allowed-tools: Read, Glob, Grep, Bash(git:*), Bash(curl:*) How it starts
The opening of the file, as written. The whole thing — 229 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Forgejo advisory review — procedure
You have been handed a kanban card that asks you to review one pull request
on git.home.claydon.co and post advisory feedback. This skill is the
procedure. Your profile carries the rails (advisory only, read-only, one
COMMENT review, git.home.claydon.co is the only host); this skill tells
you how to carry it out.
The whole job is: one Forgejo review of type COMMENT, then a completed
kanban card. Nothing else touches the PR.
0. Read the card body
The card body is a block of key: value lines:
| key | meaning |
|---|---|
repo |
owner/name on git.home.claydon.co |
pr |
PR number (<n> below) |
title |
PR title |
author |
PR author's login |
base |
base ref @ base sha |
head |
head ref @ head sha |
url |
PR web URL |
clone |
git clone URL (https://git.home.claydon.co/<owner>/<name>.git) |
changed_files |
count, or unknown |
review_skills |
comma-separated language reviewer skill names — MAY BE EMPTY |
review_skills lists language-specific reviewers only (for example
shell-script-reviewer, go-review). This workflow skill is separate and
is always loaded for you; it is never listed there.
1. Clone the repo at the head sha
Work in your scratch workspace:
git clone --depth 50 <clone> repo && cd repo
git fetch --depth 50 origin refs/pull/<n>/head
git checkout FETCH_HEAD
refs/pull/<n>/head also covers fork PRs. After checkout, git rev-parse HEAD should equal the card's head sha — if it does not, note the mismatch
in the review body and continue against what you actually checked out.
Read whole files for context. Never review from the diff alone.
If the clone or the checkout fails, kanban_block with the reason and stop.
2. Fetch the PR's unified diff
curl -fsS -H "Authorization: token $FORGEJO_TOKEN" \
"https://git.home.claydon.co/api/v1/repos/<owner>/<name>/pulls/<n>.diff"
The diff tells you which files and which lines changed. If the diff fetch
fails, kanban_block with the reason and stop.
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.
- 2d ago Changed 61d7fd94170a
- 1mo ago First seen · 229 lines · 86 tokens per session scan C 18d343c75a94
forgejo-advisory-review is a skill published in the GitHub repository rikdc/ai-skills (2 stars, last pushed 2d ago), licensed MPL-2.0. It adds 86 tokens to every session and 2,367 once invoked, about $0.0003 per session on Opus 5.5. A static security scan graded it C with 2 findings (recursive force delete, makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-04.
Other skills, from other repositories
release
Cut a brooks-lint release: set the version in package.json, propagate it across all four plugin manifests and every version-bearing text file (README badges, docs site metadata), write the CHANGELOG entry, validate, then commit, push, tag, and publish the GitHub release. Triggers when the maintainer asks to "release"…
pr
Comprehensive PR/issue review - analyzes architecture, tests, identifies unrelated changes mixed in, drafts review comment or issue comment. Use when user asks to review a PR, check a PR, look at PR changes, or comment on an issue.
pr-threads-address
Address PR review feedback by systematically working through every unresolved PR review thread on the current branch's PR - analyze each comment, make the requested code changes (with tests where useful), commit, and optionally reply and resolve.
han-release
Cut a Han release: update CHANGELOG.md with the changes since the last release, bump and tag every plugin that changed as {plugin-name}--v{version} so a version-constrained dependency can resolve, and publish a GitHub release crediting every merged pull request and closed issue to the people behind it. Use when…
pr-comment
Post friendly review comments to a GitHub PR — prepare locally, preview, then submit as atomic review. Use when: posting code review comments, giving PR feedback, sending inline suggestions. Not for: reading existing reviews (use load-pr-review), creating PRs (use create-pr), PR status (use pr-summary).
deploy-landed
A deployment verification procedure that checks whether the intended code revision is actually running, its environment settings are present, and the product works. A deployment is the process of putting new software on a server or hosting platform.