kc-rules-review

kc-rules-review is a skill for Claude Code, Codex from iamcxa/kc-claude-plugins. It costs 73 tokens per session (8,107 once invoked), scanned A, original, MIT.

A workflow for checking whether agent rules are being followed and whether each rule still earns its place. Agent rules are instructions that guide coding assistants across projects.

In plain words
What is it for?
Use it to review user and project rule files, measure rule friction and use, and change rules one decision at a time.
Why use it?
It helps find rules that are ignored, repeatedly cause corrections, or add complexity without affecting work.

Skill for Claude CodeCodex

Written for Claude Code and Codex: ${CLAUDE_PLUGIN_ROOT} variable, but also runs codex exec. Also seen: reads .claude/ paths; mentions CLAUDE.md; mentions Claude Code.

Runs only inside its plugin — its command needs a path that Claude Code sets for a plugin’s own hooks and for nothing else. Install the plugin, not this.

Part of the kc-team-ops plugin — 8 skills, 3 agents shipped together

Good fit Use it to review user and project rule files, measure rule friction and use, and change rules one decision at a time.

Compare 6 skills from other repositories ↓
Install

Getting it into your agent

This one installs as part of its plugin. Adding the marketplace and installing the plugin brings it with everything else the plugin ships.

Claude Code
/plugin marketplace add iamcxa/kc-claude-plugins
Claude Code
/plugin install kc-team-ops

Made for: Claude Code, Codex.

Or install kc-team-ops, the plugin that ships this one along with the rest of its 8 skills, 3 agents.

Wrote 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.

agentmods badge for kc-rules-review

README.md
[![agentmods](https://agentmods.dev/badge/skills/iamcxa/kc-claude-plugins/kc-rules-review/github.svg)](https://agentmods.dev/skills/iamcxa/kc-claude-plugins/kc-rules-review)
Your own site
<a href="https://agentmods.dev/skills/iamcxa/kc-claude-plugins/kc-rules-review"><img src="https://agentmods.dev/badge/skills/iamcxa/kc-claude-plugins/kc-rules-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.

agentmods 80×15 button for kc-rules-review

Your own site · 80×15
<a href="https://agentmods.dev/skills/iamcxa/kc-claude-plugins/kc-rules-review"><img src="https://agentmods.dev/badge/skills/iamcxa/kc-claude-plugins/kc-rules-review.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 73 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 8,107 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5.1 $0.00073 $0.08107
Opus 5 $0.00036 $0.04054
Sonnet 5 $0.00015 $0.01621
Haiku 4.5 $0.00007 $0.00811

Measured 6d ago against content hash ce6b479effce, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-10, from the pricing page.

Security

Grade A, and why

kc-rules-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 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.

kc-team-ops/skills/kc-rules-review/SKILL.md · 550 lines

How it starts

The opening of the file, as written. The whole thing — 550 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Rules Review

Audit an operating rule set — ~/.claude/CLAUDE.md, ~/.codex/AGENTS.md, project instruction files, memories — against what actually happened in the user's sessions, then change it one decision at a time.

Core principle: measure two numbers per rule, not one. Friction is how often the user repaired your output. Firing is how often the rule left an observable trace. The verdict lives in the pair, never in either alone.

The trap this exists to stop: friction points at one loud rule, and fixing that rule feels like finishing. Measured against a control, the loud rule gets fixed either way — what goes unasked is whether it ever ran, whether something else already owns it, and what the quiet rules' zeros mean. The quiet rules are then scored on friction alone, which is how a safety rule that has never been violated gets read as dead weight.

When to use

  • The user says some version of "you keep doing this" or "I correct the same thing every time".
  • The user quotes their own rule back at you — the strongest possible signal, because it proves the rule exists and did not fire.
  • A rule file has grown and nobody knows which parts still earn their place.
  • A rule change just landed and downstream files may quote the old section names.

Do not use for one skill's behaviour inside one repository, or for a bare inventory of stored memories. This is the whole-rule-set pass: it reads the rules that govern every session, and its unit of change is a rule, not a single correction.

Step 0 — Ask how far back, before reading anything

The window changes every number in this audit, and it is the user's call, not a default you pick for them. Ask it with the harness's question UI as the first action, offering two weeks, one month, two months, and a custom date — and say what each costs, because they find out otherwise only after waiting: a day returns in seconds, two weeks across six hundred sessions takes a few minutes, and a month across eleven hundred took eleven — long enough to exceed a ten-minute tool timeout and need backgrounding, which is worth saying before they choose it rather than after.

Read the full file on GitHub · 550 lines

Changes

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.

  1. 6d ago First seen · 550 lines · 73 tokens per session scan A ce6b479effce

Subscribe to this mod's changes

kc-rules-review is a skill published in the GitHub repository iamcxa/kc-claude-plugins (3 stars, last pushed today), licensed MIT. It adds 73 tokens to every session and 8,107 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 2026-09-03.