Menu
Use when the user needs a concise explanation of when this skill should trigger and what it does.
Install with Codex or Claude Copy this prompt, paste it into Codex, Claude, or another assistant, and let it review the skill page and install it for you.
Based on SOC occupation classification
Use when the user asks to report, summarize, audit, or present the implementation result of a plan, roadmap, PRD, engineering plan, refactor plan, migration plan, release plan, or task list. Produces evidence-based completion reports with overall, stage, and module completion percentages; goal-to-result verification tables; explicit concrete engineering benefits; unfinished-work impact analysis; and concrete next actions, using tables and charts instead of vague prose.
Use when the user asks for a software engineering plan, implementation plan, refactor plan, migration plan, rollout plan, bug-fix plan, performance optimization plan, security change plan, DevOps / CI/CD plan, technical execution plan, or review of an existing engineering plan. Produces phased, executable, independently verifiable, reviewable, rollback-aware plans with strict phase gates, plan-to-code completeness, benefit validation, and chain-state logging evidence for software engineering work.
Use when a task needs independent adversarial review during vibe coding, design work, implementation, testing, release planning, documentation, skill creation, or agent workflow design. It uses fresh internal subagents, or a user-approved local CLI reviewer fallback when internal subagents are unavailable, to attack artifact assumptions, happy paths, implementation completeness, target benefit realization, user-perspective usability, failure scenarios, and evidence gaps through formal /vs_review/ reports and mandatory response closure.
Use when Codex needs to design, reorganize, audit, or maintain a repository's documentation system: docs/ directory hierarchy, global vs version-level documents, version PRDs, technical designs, engineering plans, document versioning, naming conventions, link/index management, archive policy, or automated docs structure validation.
Use when debugging medium or large bugs, regressions, flaky failures, production symptoms, repeated failed fixes, or user-explicit requests where the agent must combine persistent /coe case tracking, independent multi-path root-cause investigation, and a hard diagnostic evidence gate before repair design or implementation.
Use when a user brings a new product, feature, app, website, internal tool, automation, workflow, or other product requirement that needs multi-round clarification before implementation. The skill keeps the conversation focused on product logic, user experience, interaction design, rules, edge cases, and acceptance criteria, then writes a Draft or Ready PRD.
| name | your-skill-name |
| description | Use when the user needs a concise explanation of when this skill should trigger and what it does. |
State what this skill helps with and what good output looks like.
references/ files only when they materially improve results.scripts/ only when deterministic execution is better than re-writing code in-context.