Evaluator Library — Security skill for Claude Code
Dispatch a fresh-context evaluator against a domain rubric.
How to install Evaluator Library
Installs to ~/.claude/skills/adrijshikhar-catalyst-evaluator-library/SKILL.md
mkdir -p ~/.claude/skills/adrijshikhar-catalyst-evaluator-library && curl -fsSL https://raw.githubusercontent.com/adrijshikhar/catalyst/HEAD/commands/evaluator-library.md -o ~/.claude/skills/adrijshikhar-catalyst-evaluator-library/SKILL.md Restart Claude Code, or start a new session, for it to be picked up.
What Evaluator Library does
description: Dispatch a fresh-context evaluator against a domain rubric. Bundled domains: code-quality, ui-design, prose, security, performance, accessibility. Each dispatch enforces the anti-self-grade rule. Use to score an artifact when PIPELINE mode reaches an evaluator stage, or as a standalone quality gate.
Invoke the `evaluator-library` skill.
Recognized sub-commands (parse `$ARGUMENT`):
list— List the available domains. Readskills/evaluator-library/evaluators/*.md(bund
Alternatives in Security
- Release Audit — Use right before cutting a release — it spawns ONE fresh sibling session via CCC that audits only the JUDGMENT 130 ★
- Audit Ddd — DDD architecture audit with pattern recommendations 63 ★
- Sc Analyze — Comprehensive code analysis across quality, security, performance, and architecture domains 55 ★
Full documentation available on GitHub
View Source RepositoryRelated Skills
Evaluator
Use when the user invokes /evo:evaluator or asks to evaluate, assess, score, or independently audit code quali
Academic Register
Claude Code skill: audit & fix English academic prose register — guards BOTH poles (too-informal ↔ AI-inflated
Skill Detail
Shared model-rubric detail + plan-doc update pattern for gate-a~e, audit, doc-cleanup. Load on demand (not aut
V Archaeology
Run Compound V Phase 1A — code archaeology — on a topic without running the other pre-flights. Useful when you
Gaia Harden
Judge-the-form, human-gated hardening. Reviews recurring code-audit-frontend findings and, with approval, draf
Pbi Audit
Run a comprehensive 21-rule health check across 8 domains: relationships, naming, date table, measure quality,
Related Agents
Martha Judge
Strict, independent evaluator for Martha's sales copy. Use only inside Martha's improvement cycle (IMPROVE.md)
Harness Checker
Fresh-context artifact evaluator for goal loops. Reads only final artifacts — never Maker reasoning or PROGRES
Clio
Scribe. Writes notes and decision records to the brain paths only, with a strict anti-fabrication rule: every