Skill Claude CodeCodex
Draft a feature-level specification at doc/specs/NNNN- .md following the six-layer artifact stack (Constitution → Domain → Product → Spec → Plan/Decisions → Code). Adapts GitHub Spec Kit's mandatory sections (User Scenarios, Requirements, Success Criteria) to the kit's documentation discipline. Use when the user wants…
Skill Claude CodeCodex
Scaffold a staged spike with golden fixtures per WORKFLOW.md §14, for cases where the spec is clear but the technique is uncertain across multiple plausible approaches. Four stages — discovery, golden fixture, pipeline with gates, two-layer evaluation. Use when the unknown is how, not what. Triggers on "spike"…
Skill Claude CodeCodex
Draft a new Codex custom subagent at .codex/agents/ .toml, using the official Codex subagents format. Use when the user wants to create, write, draft, or scaffold a custom Codex subagent for delegated work (fresh-context reviewer, codebase explorer, docs researcher, test designer, bug reproducer, bounded…
Skill Claude CodeCodex
Draft a new task tracking file at doc/tasks/NNNN- .md, using a checkbox-toggle + append-only-Notes format optimized for LLM editing. Use when the user wants to create, draft, scaffold, or open a task, ticket, work item, or backlog entry tracked in the repo. Status starts at proposed.
Skill Claude CodeCodex
Test-Driven Development per WORKFLOW.md §16 — red-green-refactor as deterministic LLM guardrail. Five phases — confirm regime, plan vertically, tracer bullet (one test → red → minimum code → green), incremental loop, refactor while green. Tests verify behavior through public interfaces; horizontal slicing (bulk-write…
Skill Claude CodeCodex
Outcome-based prompting per WORKFLOW.md §9. Give the agent the finish line first, not the path. Five steps — confirm regime, ground truth pair, Test Dependency Map, three approaches, pick by one criterion, implement and verify. Triggers on "outcome-based", "TDG", "ground truth", "expected output", "three approaches"…
Agent Claude Code
Adversarial fresh-context reviewer for one rule-group of an ad-audit maximum-gate audit. Reads a handoff carrying a single group's rules plus the target, walks those rules as a literal checklist, and emits an explicit verdict for every rule — grounding on the actual code/output, citing the exact rule, and demanding an…
Agent Claude Code
Adversarial fresh-context code reviewer per WORKFLOW §10. Reads an axis-bounded handoff (diff + the slice of context relevant to one axis only) and reports findings strictly within that axis. The caller chooses the axis — Standards (does the diff conform to AGENTS.md / ARCHITECTURE.md / GUIDELINES.md / CONTEXT.md /…
Agent Claude Code
Adversarial fresh-context reviewer for one proposed rule-set candidate in an ad-level-up curation pass. Reads the candidate plus the actual current rule files and tries to REFUTE it along a single assigned lens — already-covered, coherence-and-necessity, or placement — never assuming, always citing the covering rule…
Instructions file CodexOpenCode
Instructions for CorridorTech/PoseCap, covering agents.md, project overview, setup, build, test, install (engine bridge + dev tooling) and test (single file preferred over full suite).