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 agentmods add skills/florianbruniaux/claude-code-plugins/eval-rulesnpx skills add FlorianBruniaux/claude-code-plugins --skill eval-rulesgit clone --depth 1 https://github.com/FlorianBruniaux/claude-code-pluginsWrote 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/florianbruniaux/claude-code-plugins/eval-rules)<a href="https://agentmods.dev/skills/florianbruniaux/claude-code-plugins/eval-rules"><img src="https://agentmods.dev/badge/skills/florianbruniaux/claude-code-plugins/eval-rules.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 | $0.00084 | $0.02197 |
| Opus 5 | $0.00042 | $0.01099 |
| Sonnet 5 | $0.00017 | $0.00439 |
| Haiku 4.5 | $0.00008 | $0.00220 |
Grade A, and why
eval-rules 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 yesterday.
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 — 226 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Rules Evaluator
Discover all rule files, validate their structure and glob patterns against the real project, then run an interactive session to confirm (or improve) each rule with the user.
The goal is not just to score; it is to leave the rules directory in better shape than it was.
When to Use
- First time writing
.claude/rules/files (validate before committing) - A rule seems to never trigger, or fires on every file
- Migrating
@imports from CLAUDE.md to path-scoped rules - Periodic hygiene: "are these rules still relevant to how we work?"
- After onboarding to a new codebase
Key Concepts
| Mechanism | When it loads | Notes |
|---|---|---|
@file in CLAUDE.md |
Session start, always | Even inside a conditional sentence |
No paths: in rule |
Session start, always | Same cost as @import |
paths: frontmatter |
When Claude reads a matching file | Trigger = Read tool on a matched file |
User-level rule (~/.claude/rules/) |
Session start, always | Applies to all projects on the machine |
The paths: field is the main lever for keeping rules contextual. An always-on rule with 80 lines loads on every session even if you're fixing a typo in README.md.
Glob patterns in paths: support brace expansion: "src/**/*.{ts,tsx}" matches both .ts and .tsx files with one entry.
Scoring Criteria (12 pts per rule)
| # | Criterion | Max | What is checked |
|---|---|---|---|
| 1 | frontmatter block | 1 | File has YAML frontmatter (--- delimited) |
| 2 | paths: field | 2 | Present (1pt) + at least one pattern listed (1pt) |
| 3 | pattern validity | 3 | Each pattern matches >= 1 file in project (up to 3 patterns checked) |
| 4 | scope | 2 | Not dead (>= 1 match) + not too broad (<30% of project source files) |
| 5 | content quality | 3 | Has clear header/title (1pt) + rules are specific/actionable (1pt) + under 150 lines (1pt) |
| Bonus | focus | +1 | Under 15 rules in file |
Thresholds:
- Good: >= 10/12 (>= 83%)
- Needs work: 7-9/12 (58-82%)
- Fix: < 7/12 (< 58%)
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.
- yesterday First seen · 226 lines · 84 tokens per session scan A 0f13fe736dac
eval-rules is a skill published in the GitHub repository FlorianBruniaux/claude-code-plugins (40 stars, last pushed 3d ago), licensed MIT. It adds 84 tokens to every session and 2,197 once invoked, about $0.0004 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 2026-09-04.
Other skills, from other repositories
vc-autopilot
Emit and validate the provisional goal block for Autopilot Mode. Owns the 9-field format and resume detection from a pasted goal block.
vc-problem-solving
Apply systematic problem-solving techniques when stuck. Use for complexity spirals, innovation blocks, recurring patterns, assumption constraints, simplification cascades, scale uncertainty.
thoroughness-scoring
Score every decision point with a Thoroughness Rating (1-10). AI makes the marginal cost of doing things properly near-zero — pick the higher-rated option every time. Includes scope checks to distinguish contained vs unbounded work.
report
Read the delivery log and say which rules actually fire, which never have, and what to prune or fix. Use when the user asks whether ballast is doing anything, wants to clean up their rule catalog, or on a periodic review.
orchestrate
Wire Commands, Agents, and Skills together for complex features. Use when building features that need research, planning, and implementation phases.
squid-self-improve
Analyze developer corrections from the current coding session and persist lessons learned as rules in AGENTS.md files or memory. Use at the end of a session after the developer corrected your work, when they say "squid-self-improve", ask to capture what was learned, or ask you to reflect on mistakes and extract…