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 commands/sdsrss/claudemd/claudemd-rulesgit clone --depth 1 https://github.com/sdsrss/claudemdWrote 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/commands/sdsrss/claudemd/claudemd-rules)<a href="https://agentmods.dev/commands/sdsrss/claudemd/claudemd-rules"><img src="https://agentmods.dev/badge/commands/sdsrss/claudemd/claudemd-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 | $0.00067 | $0.00629 |
| Opus 5 | $0.00034 | $0.00315 |
| Sonnet 5 | $0.00013 | $0.00126 |
| Haiku 4.5 | $0.00007 | $0.00063 |
Grade A, and why
claudemd-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 4d 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.
What it actually says
Default window is 30 days (per §0.1 v6.11.15 — lowered from 90d because the 90d gate was structurally unreachable under typical rule-hits log retention; with 30d the audit can actually produce demote candidates instead of always reporting insufficientData).
$ARGS parsing: split $ARGS into a numeric day-count and a --verbose agent-presentation flag (NOT a script flag — the script ignores it):
$ARGS value |
CLAUDEMD_RULES_DAYS env |
Agent output |
|---|---|---|
| (empty) | unset (script uses default 30) | summary only |
90 |
90 |
summary only |
--verbose |
unset | summary + full rules array |
90 --verbose |
90 |
summary + full rules array |
Passing --verbose (or any non-numeric token) directly to CLAUDEMD_RULES_DAYS would crash the script with --days requires a positive integer (got '--verbose') — always strip the flag before setting env.
Run: CLAUDEMD_RULES_DAYS=<numeric-from-$ARGS-or-30> node ${CLAUDE_PLUGIN_ROOT}/scripts/hard-rules-audit.js
The JSON contains:
| Field | Meaning |
|---|---|
byScope |
core vs extended HARD-rule counts (§13.1 size budget cross-check) |
byEnforcement |
hook / self / external / both — which rules are mechanically gated vs agent self-discipline |
byConfidence |
high / medium / low — author's pre-data assessment of rule reliability |
demoteCandidates |
hook-enforced rules with 0 hits in window — §13.1 demotion review queue |
staleReviews |
rules whose last_demote_review is null, unparseable, or older than the §13.1 review cadence (28d, per OPERATOR.md) — operator's demote-review queue. Independent of --days, which only sets the hit-counting window. |
rules |
per-rule rows with hits: {total, deny, bypass, warn} (null for self/external — agent text not yet captured; R-N8 transcript scan is the fix path) |
Format: surface byEnforcement + byConfidence summary first, then list demoteCandidates and staleReviews as action items. Suppress full rules array unless $ARGS contained --verbose (per the parsing table above).
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.
- 4d ago First seen · 33 lines · 67 tokens per session scan A 957e11eaa104
claudemd-rules is a command published in the GitHub repository sdsrss/claudemd (2 stars, last pushed 10d ago), licensed MIT. It adds 67 tokens to every session and 629 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 2026-08-31.
Other commands, from other repositories
brooks-debt
Run a Brooks-Lint tech debt assessment.
brooks-health
Run a Brooks-Lint codebase health dashboard across all four dimensions.
plan
Turn an approved spec into an implementation plan an engineer with zero context could execute — with a quality controller that blocks placeholders and hollow tasks.
simplify
The over-engineering review: five tags (delete, stdlib, native, yagni, shrink), a mandatory replacement per finding, and a real null result when there is nothing to cut.
audit
Onboard an existing codebase: every domain's checks over the whole tree, then a triaged plan to bring it in line.
git
The pre-finish status: branch, hygiene findings, message checks, workflow lint, template state.