park-jun-woo

Full-stack SSOT orchestrator that validates the consistency of 10 declarative sources (features, manifest, OpenAPI, SQL DDL, sqlc, SSaC, Mermaid stateDiagram, OPA Rego, Hurl, STML) and generates Go+Gin backend and React frontend code from them. Use this skill when writing, editing, or validating SSOT spec files for a yongol project, when troubleshooting cross-layer validation errors, or when generating backend/frontend code from declarative specifications.

Build a quest CLI in Go with the reins framework — it moves the authority to declare "done" from the AI to a deterministic machine gate ("generation is probabilistic; verification is deterministic"). Use this skill when asked to build a quest CLI, a reins consumer, a generate→verify→retry loop, a deterministic completion gate, or an unattended LLM generation loop whose output a machine judges. Plug in one domain contract (gate.Definition) and reins supplies the ratchet, command skeleton, aggregation, feedback, and export; optionally move judgment into a runtime-interpreted TANGEUL gate document (gate.md) via GateOptions. Triggers on keywords like reins, quest CLI, gate.Definition, gate.Rule, deterministic gate, quest-CLI, unattended loop, TANGEUL, gate.md, or turn model.

Test coverage ratchet for Go, TypeScript, Python, Rust, Java, and C#. Built on the reins deterministic quest gate, indexes every function, detects missing tests, measures branch coverage, and drives LLM agents to fill gaps — one function at a time or in an unattended loop. Use this skill when writing unit tests for legacy codebases, measuring test coverage, or running test-generation loops with LLM agents. Triggers on tasks involving tsma commands, test coverage improvement, or missing test detection.

Static analysis tool that extracts OpenAPI specs, DDL schemas, and SQL skeletons from 16 web frameworks (Go+Gin, Fiber, Echo, NestJS, Fastify, Hono, FastAPI, Flask, Django, Express, Spring Boot, Quarkus, ASP.NET Core, Supabase Edge Functions, Actix Web, Laravel). Use when you need to generate or update API specs, extract endpoint metadata, manage DDL migrations, scaffold sqlc queries, or merge scan results with existing OpenAPI specs.

Ratchet tool for wall-to-wall Hurl API test generation. Use when writing Hurl tests for SaaS backends, when an endpoint lacks test coverage, or when asked to generate API integration tests. Triggers on keywords: hurl, endpoint test, API test, ratchet, coverage.

Operate and set up an agent-run blog. Use when working in a repo with a blog.yaml SSOT — scaffolding a blog (abloq init), generating derived files (hugo.toml/robots.txt/llms.txt/JSON-LD), running GEO/structure/evidence gates, or executing the writing/translation/refresh/evidence/cluster quests. Triggers on blog.yaml, insight.yaml, quests/queue, "abloq gate", citation/source verification, hreflang/freshness/cluster checks, and self-hosting the abloqd backend.

Code structure convention and CLI for LLM-native Go/Python/TypeScript/C#/Java/Rust development. Use when: validating "one file, one concept" rules, tracing func call chains, exploring code context with LLM scoring, verifying annotation-body consistency. Triggers on: filefunc, code structure validation, annotation rules, call chain, func tracing, codebook, one file one func, ff:func, ff:what, ff:type.