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 cajasmota/grafel --skill grafel-aware-reviewgit clone --depth 1 https://github.com/cajasmota/grafelWrote 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/cajasmota/grafel/grafel-aware-review)<a href="https://agentmods.dev/skills/cajasmota/grafel/grafel-aware-review"><img src="https://agentmods.dev/badge/skills/cajasmota/grafel/grafel-aware-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/cajasmota/grafel/grafel-aware-review"><img src="https://agentmods.dev/badge/skills/cajasmota/grafel/grafel-aware-review.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
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 | $0.00067 | $0.05615 |
| Opus 5 | $0.00034 | $0.02808 |
| Sonnet 5 | $0.00013 | $0.01123 |
| Haiku 4.5 | $0.00007 | $0.00562 |
Grade A, and why
grafel-aware-review 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 11d 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 — 635 lines — stays where its author put it; the contents beside it link to each section on GitHub.
grafel-aware-review
A structural review protocol for AI agents. This skill turns grafel MCP into a systematic second opinion on every PR — one that catches impact leakage, broken call chains, and pattern drift that diff-only review misses.
Prerequisite: the /using-grafel skill. This skill assumes you already
know the 22 MCP tools and the orientation workflow.
1. Should I run grafel on this PR?
Yes — run the structural checks when the diff touches:
| Change type | Key risk grafel catches |
|---|---|
| New/renamed function or method | Callers in other modules still expect the old signature |
| Deleted function or method | Orphan call-sites remain in any repo of the group |
| New HTTP endpoint | No client calls it yet (dead definition); or it clashes with an existing route |
| Changed HTTP endpoint path/method | Existing client calls become orphan callers |
| New inter-module import | Introduces an unexpected cluster coupling or a cycle |
| New message publish/subscribe | No subscriber (dead publish) or no publisher (dead subscriber) |
| Cross-repo call site added | Target entity may not exist in the other repo's graph |
| Pattern implementation (new handler, new serializer, etc.) | Does not follow the established structural pattern |
No — skip grafel when the diff is limited to:
- Whitespace, formatting, or lint fixes.
- Comment or docstring edits.
- Test fixture data files (JSON, YAML, CSV).
- Changes that are entirely within a single private function with no exported signature.
- Config-only changes (
*.env,*.yaml,*.toml) with no code entry points. - Pure CSS/markup with no JS/TS wiring changes.
Quick self-check: if the change cannot be reached from any entity in the grafel graph (no exported name, no route, no published event), skip the structural pass and review the diff purely on its own terms.
2. Quick win — one call for 80 % of insights
When time is short, a single call surfaces most structural risk:
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.
- 11d ago First seen · 635 lines · 67 tokens per session scan A 13bcb02a2137
grafel-aware-review is a skill published in the GitHub repository cajasmota/grafel (15 stars, last pushed 2d ago), licensed MIT. It adds 67 tokens to every session and 5,615 once invoked, about $0.0003 per session on Opus 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
autoreview
Pre-commit/ship code review: Codex default; optional Claude or Pi.
omh-code-review
This is a Hermes-native code-review workflow skill.
revdiff-plan
Review the last Codex assistant message (plan, analysis, or proposal) with inline annotations in a TUI overlay. Extracts the most recent response from Codex rollout files and opens it in revdiff for review and annotation. Activates on "revdiff-plan", "review plan with revdiff", "annotate plan", "review last response"…
code-reviewer
Code review specialist focused on patterns, bugs, security, and performance.
full-repo-review
Comprehensive four-wave review of all repo source files, producing a prioritized issue backlog.
agent-teams-simplify-and-harden
Implementation + audit loop using parallel agent teams with structured simplify, harden, and document passes. Spawns implementation agents to do the work, then audit agents to find complexity, security gaps, and spec deviations, then loops until code compiles cleanly, all tests pass, and auditors find zero issues or…