architect
25Agent
Inspects: API contract, cross-project, arch adherence.
Agent
Inspects: API contract, cross-project, arch adherence.
Agent
Inspects: AC pass, user scenarios, edge cases, UX.
Skill Claude CodeCodex
Create workspace-owned PRD or technical-design contracts. Stay within the docs scope; never hand off, implement, commit, push, or open an MR from this skill.
Skill Claude CodeCodex
A coordinator for taking a software ticket from requirements through design, implementation, review, and merge request. It uses quality gates, recorded documents, approvals, and validation between stages.
Plugin Claude Code
Audit Claude skills and commands for publish readiness with 6-category rubric scoring and concrete fixes.
Skill Claude CodeCodex
Use when the user explicitly asks to audit or quality-check Claude skills/commands for publish readiness (for example: 'skill audit', 'audit this skill', 'check skill quality'). Scores against Anthropic skill best practices and returns concrete fixes. Do not use for general code review, document polishing, or feature…
Plugin Claude Code
Pokemon Pokedex-style skill dashboard with usage tracking, leveling system, achievements, and visual analytics.
Skill Claude CodeCodex
Pokemon Pokedex-style skill dashboard. Tracks usage, levels, rarity, achievements. Use when: 'show skills', 'skill stats', 'open dashboard'. Not for skill auditing.
Plugin Claude Code
깐깐징어 — 5-axis MR/PR reviewer (security / blast radius / tests / quality / observability) with a repo manifest system, deterministic rules registry pre-pass, and optional axes (a11y, intent-alignment, lifecycle-necessity, doc-review).
Skill Claude CodeCodex
A strict code-review skill that examines merge requests and pull requests across security, change impact, tests, code quality, and observability, with optional checks such as accessibility and rule compliance.
Plugin Claude Code
Evidence-based QA and verification: risk-based scenario discovery, an evidence rigor ladder with strict PASS/FAIL/BLOCKED verdicts, checklist-driven E2E with human gates, and competing-hypothesis debugging.
Skill Claude CodeCodex
A tool for running a prepared end-to-end checklist against the real running application. End-to-end testing follows a complete user journey, such as opening a page, submitting a form, and checking the result, while recording evidence.
Skill Claude CodeCodex
A structured method for debugging failures by comparing multiple possible causes and testing which ones the evidence rules out. It examines code, the environment, data, and measurement errors.
Skill Claude CodeCodex
A risk-based testing skill for checking implemented features, branches, APIs, and screens, including unusual user actions and edge cases. It requires observable evidence before marking a test as passing.