By edonghyun
Generate, drift-check, lint, test-stub, coverage-audit and scope-walk E2E specs against the current implementation. A six-skill safety net for keeping code, spec, tests, and team knowledge in sync.
Drift-check the current implementation against the existing specs/ directory. Use when the user wants to verify the spec is still accurate, catch undocumented changes, or run the safety-net check after recent code edits. Triggers include "spec compare", "drift check", "verify spec", "is the spec stale", "check spec drift", "스펙 비교", "스펙 드리프트", "스펙 확인", "스펙 검증". Produces specs/DRIFT.md categorizing added / removed / changed elements per layer and per flow, with severity ratings. Does NOT modify the existing specs — only reports.
Cross-reference existing tests against the specs/ directory to find untested flows, endpoints, and invariants. Use when the user wants to see how much of the spec is actually covered by tests, before relying on the test suite as a safety net. Triggers include "spec coverage", "test coverage vs spec", "what's untested", "coverage report", "스펙 커버리지", "테스트 커버리지", "안 테스트된 거 뭐야". Produces specs/COVERAGE.md. Never modifies existing tests or specs.
Generate test stubs from the existing specs/ directory and record whether they are active todos or inactive planning stubs. Use when the user wants to turn the spec into planned executable coverage for domain invariants, backend endpoints, and e2e flows. Triggers include "gen tests", "generate tests from spec", "spec to tests", "build test stubs", "테스트 생성", "스펙 기반 테스트". Detects the project's existing test framework. Writes stubs only — never overwrites existing tests. Asks the user for confirmation if more than 20 stubs would be created.
Generate end-to-end specification markdown files that mirror the current implementation as a safety net. Use when the user wants to extract, document, or snapshot the app's behavior — across frontend action flows, backend use cases, queries, and domain features. Triggers include "generate spec", "spec mirror", "e2e spec", "document app", "mirror implementation", "스펙 생성", "스펙 미러", "구현 스펙 만들어줘", "전체 스펙 정리". Produces layered specs (frontend / backend / domain) plus flow specs that cross-reference layers. Asks the user for a brief architecture confirmation after its own analysis, before writing files.
Audit the specs/ directory for internal hygiene issues — dead cross-references, missing source lines, stale [INFERENCE] markers, orphaned KEEP blocks, broken refs/ links. Use when the user wants to verify the spec itself is healthy, after manual edits to spec files, or before relying on the spec as a safety net. Triggers include "lint spec", "spec hygiene", "spec audit", "check spec health", "스펙 린트", "스펙 점검", "스펙 정합성". Produces specs/LINT.md and never modifies any other file. Different from compare (which checks spec vs code) — lint checks spec vs spec.
Own this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimOwn this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimBased on adoption, maintenance, documentation, and repository signals. Not a security audit or endorsement.
A host-neutral personal plugin marketplace for coding agents, focused on safety nets — plugins that keep code, spec, tests, and team knowledge in sync rather than generating net-new content.
Read
docs/PHILOSOPHY.mdfor the mission,docs/CONVENTIONS.mdfor the rules every plugin follows, anddocs/CONTRIBUTING.mdto add a new plugin or skill.
# Add marketplace from GitHub.
claude plugin marketplace add edonghyun/donghyuns-agent-tools
# Or, during local development.
claude plugin marketplace add /Users/idonghyeon/projects/donghyuns-agent-tools
# Install plugins.
claude plugin install spec-mirror@donghyuns-agent-tools
claude plugin install qa-operator@donghyuns-agent-tools
claude plugin install feature-intake@donghyuns-agent-tools
# Refresh after marketplace changes.
claude plugin marketplace update donghyuns-agent-tools
claude plugin update qa-operator@donghyuns-agent-tools
claude plugin update feature-intake@donghyuns-agent-tools
Codex support reuses the same plugins/<plugin>/skills/ tree and adds Codex-specific manifests next to the Claude manifests:
.agents/plugins/marketplace.jsonplugins/<plugin>/.codex-plugin/plugin.json# Add marketplace from GitHub.
codex plugin marketplace add edonghyun/donghyuns-agent-tools --ref main
# Or, during local development.
codex plugin marketplace add /Users/idonghyeon/projects/donghyuns-agent-tools
# Install Codex-supported plugins.
codex plugin add qa-operator@donghyuns-agent-tools
codex plugin add feature-intake@donghyuns-agent-tools
# Refresh after marketplace changes.
codex plugin marketplace upgrade donghyuns-agent-tools
codex plugin add qa-operator@donghyuns-agent-tools
codex plugin add feature-intake@donghyuns-agent-tools
Start a new Codex thread after install or update so the refreshed plugin skills are loaded into the session.
Turn external feature artifacts into product integration analysis packages. feature-intake takes VoC-driven PoCs, demo apps, Figma prototypes, internal tools, AI experiments, or screenshots from another team and produces screenshot-backed user flows, user journeys, use cases, product framing, integration fit, risks, MVP boundaries, and handoff documentation.
Just say:
이 PoC를 feature-intake 해줘. 스크린샷 포함해서 유저플로우, 유저저니, 유즈케이스, 제품 편입 분석까지 정리해줘.
Skills:
| Command | Description |
|---|---|
/feature-intake:start | Natural-language entry. Runs inspect, capture, framing, analysis, mapping, and packaging. |
/feature-intake:inspect | Discover artifact structure, pages, controls, external calls, and capture targets. |
/feature-intake:capture | Capture page, interaction, modal/dialog, edge-case, and contact-sheet screenshots. |
/feature-intake:frame | Generalize product framing: actors, placement, output, quality bar, MVP, data/AI/storage, operations, risks. |
/feature-intake:analyze | Write user flows, journeys, use cases, inventories, and screen-by-screen annotations. |
/feature-intake:map | Map the artifact into the existing product's IA, modules, data, APIs, permissions, storage, and backlog. |
/feature-intake:package | Assemble the final screenshot-backed handoff README and coverage matrix. |
Workflow:
/feature-intake:start
├─ inspect
├─ capture
├─ frame
├─ analyze
├─ map
└─ package
Plugin docs: CONCEPTS · WORKFLOW-GUIDE · EXAMPLES · TROUBLESHOOTING
Delegate browser QA from natural language. qa-operator turns requirement lists, spreadsheets, issues, PRs, or feature ranges into a QA plan, starts a live monitor dashboard, executes browser checks, captures screenshots/logs/traces, flags UI/UX risks, triages failures, and can repair obvious local defects when allowed.
Just say:
이 시트 31~43번 QA 맡겨줘. monitor 켜고 UI/UX 문제까지 봐줘.
Skills:
npx claudepluginhub edonghyun/donghyuns-agent-tools --plugin spec-mirrorEvidence-backed UI iteration operator for comparing baseline and current web, mobile web, native mobile, and local session-derived UI states during product UI changes.
Delegated browser QA operator: plan, run, monitor, triage, and optionally repair acceptance checks from requirements with evidence dashboards.
Turn external feature artifacts, PoCs, demos, prototypes, and VoC experiments into screenshot-backed page-walk product integration analysis packages.
Comprehensive skill pack with 66 specialized skills for full-stack developers: 12 language experts (Python, TypeScript, Go, Rust, C++, Swift, Kotlin, C#, PHP, Java, SQL, JavaScript), 10 backend frameworks, 6 frontend/mobile, plus infrastructure, DevOps, security, and testing. Features progressive disclosure architecture for 50% faster loading.
Tools to maintain and improve CLAUDE.md files - audit quality, capture session learnings, and keep project memory current.
Develop, test, build, and deploy Godot 4.x games with Claude Code. Includes GdUnit4 testing, web/desktop exports, CI/CD pipelines, and deployment to Vercel/GitHub Pages/itch.io.
A growing collection of Claude-compatible academic workflow bundles. Covers scientific figures, manuscript writing and polishing, reviewer assessment, citation retrieval, data availability, paper reading, literature search, response letters, paper-to-PPTX conversion, and evidence-grounded Chinese invention patent drafting. Rules are organized as reusable skill folders with explicit workflows and quality checks.
Create new skills, improve existing skills, and measure skill performance. Use when users want to create a skill from scratch, update or optimize an existing skill, run evals to test a skill, or benchmark skill performance with variance analysis.
Unity Development Toolkit - Expert agents for scripting/refactoring/optimization, script templates, and Agent Skills for Unity C# development