Plugin Auditor — Security agent for Claude Code
Launched by /aidex:aidex audit; not for direct use.
How to install Plugin Auditor
Installs to ~/.claude/agents/yacb2-aidex-plugin-auditor.md
mkdir -p ~/.claude/agents && curl -fsSL https://raw.githubusercontent.com/yacb2/aidex/HEAD/agents/plugin-auditor.md -o ~/.claude/agents/yacb2-aidex-plugin-auditor.md Restart Claude Code, or start a new session, for it to be picked up.
What Plugin Auditor does
name: plugin-auditor description: Launched by /aidex:aidex audit; not for direct use. Returns uninstall candidates by subagent cost vs usage. model: haiku effort: high tools: Read, Grep, Glob, Bash
You audit installed plugins for token cost vs. usage.
Inputs
- The snapshot JSON (
context-snapshot.py, path in the prompt):categories['custom-agents']is the measured idle cost of every plugin subagent together;skills{}carries each plugin skill'suses/last_usedwhen `u
Alternatives in Security
- Token Auditor — Scans ui/src/ for hardcoded visual values, duplicate components, and shadcn replacement candidates; produces d 79.4k ★
- Evaluator Physical Agent — Provider-neutral physical-security evaluator — plans bounded coverage and records opaque evidence references w 97 ★
- Security Code PR Reviewer — Subagent that reviews PR diffs for security, authentication, and authorization 93 ★
Full documentation available on GitHub
View Source RepositoryRelated Agents
Conventions Auditor
Launched by /aidex:aidex audit; not for direct use. Returns validate.py violations as aidex findings.
Freshness Checker
Launched by /aidex:aidex audit; not for direct use. Returns stale documentation artifacts.
Figma Analytics
Use to audit the Figma library side of the design system: override hotspots (instances drifted from their main
API Auditor
API security specialist. Launched by the audit orchestrator with assigned coverage units and surface-map hotsp
Iac Auditor
IaC and container security specialist. Launched by the audit orchestrator with assigned coverage units and sur
Balthasar
BALTHASAR-2, the mother. Judges a proposition on safety, security, and who gets hurt when it fails. Returns a
Related Skills
Token Review
Analyze SDLC token usage from .claude/sdlc/token-log.json (last run) and token-history.jsonl (rolling). Surfac
Kb Check Drift
Check whether .kb/ is stale — read-only, returns refresh candidates
Loongsuite Pilot
Local-first telemetry collector for AI coding agents — unified OpenTelemetry events for Claude Code, Codex, Cu