Skill Claude CodeCodex
Convert "done" into checkable CodeFlowMu acceptance criteria. Use before PM dispatch, PM review, QA verification, or ADMIN closure to define functional, file, behavior, evidence, and forbidden-item checks.
Skill Claude CodeCodex
Convert "done" into checkable CodeFlowMu acceptance criteria. Use before PM dispatch, PM review, QA verification, or ADMIN closure to define functional, file, behavior, evidence, and forbidden-item checks.
Skill Claude CodeCodex
Review CodeFlowMu/FCoP proposals for architecture and governance boundary risk. Use when adding tools, skills, runtime logic, Panel buttons, lifecycle behavior, report/eval/issue-draft behavior, or anything that might mix MCP Skill with Agent Playbook Skill.
Skill Claude CodeCodex
Split CodeFlowMu requirements into P0/P1/P2 delivery plans with owners, file scope, acceptance criteria, risks, and dependencies. Use when PM needs to dispatch work to OPS/DEV/QA/EVAL while preserving scope and FCoP boundaries.
Skill Claude CodeCodex
Discover, learn, validate, and retain capabilities that the PM needs but does not yet have. Use whenever a request needs unfamiliar domain knowledge, a missing tool or workflow, a new MCP capability, a new Skill, or stronger evidence than the PM can currently produce.
Skill Claude CodeCodex
Audit long, complex, and potentially inconsistent PM taskbooks and compile them into one self-contained, semantically validated Product Brief with a requirement ledger, WP dependency graph, budgets, absolute schedule, gates, tests, experiment evidence, risks, recovery, and stop conditions. Use for long-horizon or…
Skill Claude CodeCodex
Classify CodeFlowMu tasks, risks, blocked reports, and EVAL findings as P0/P1/P2. Use when PM must decide order, owner, urgency, or whether work should be handled now, later, or held for ADMIN.
Skill Claude CodeCodex
Create a PM-owned product design brief before dispatching product, UI, PWA, mobile, Gateway, app merge, or feature upgrade work. Use when PM must define positioning, user value, information architecture, interaction, visual direction, acceptance, and multi-agent split before DEV/QA/OPS execution.
Skill Claude CodeCodex
Turn vague ADMIN requests, chat feedback, screenshots, and workflow complaints into scoped CodeFlowMu product requirements. Use when PM needs PRD-style clarification, user stories, goals, non-goals, constraints, risks, or task input while preserving FCoP boundaries.
Skill Claude CodeCodex
Control CodeFlowMu task scope and prevent drive-by expansion. Use when PM must define allowed files, forbidden files, non-goals, risk expansion points, or when an agent proposes refactors, runtime changes, protocol changes, or external skill imports beyond the task.
Skill Claude CodeCodex
Analyze ambiguous product and engineering problems before action. Use for every PM request, decision, plan, diagnosis, architecture choice, UI/UX question, delivery trade-off, or task that might otherwise be delegated without a defensible solution.
Skill Claude CodeCodex
Use when PM writes or reviews development tasks and must choose the smallest appropriate technology stack, language, runtime, environment, deliverable format, and escalation boundary. Applies especially when deciding between TypeScript, Python, static HTML/CSS/JS, existing project stack, scripts, games, prototypes…
Skill Claude CodeCodex
Check plausible regressions from a CodeFlowMu change. Use when QA needs to identify impacted workflows, run focused checks, and record gaps without expanding into full-system audit.
Skill Claude CodeCodex
Reproduce CodeFlowMu bugs or reported behavior. Use when QA needs to capture environment, steps, actual result, expected result, and reproducibility before fix verification.
Skill Claude CodeCodex
Verify CodeFlowMu fixes against acceptance criteria. Use when QA needs to run commands or UI checks, record pass/fail evidence, and report remaining issues.
Skill Claude CodeCodex
Use when an agent needs to run a local command for verification, diagnostics, tests, builds, typechecks, scripts, or CLI behavior and must report truthful command evidence.
Skill Claude CodeCodex
Use when an agent needs to choose appropriate verification scope after a change, balancing focused tests, package tests, typechecks, UI checks, and known residual risk.
Skill Claude CodeCodex
Coordinate technical delivery across CodeFlowMu roles without taking ADMIN authority. Use when a technical manager needs to split workstreams, sequence dependencies, define handoff points, track technical risks, or coordinate DEV/QA/OPS/PM/EVAL delivery.
Skill Claude CodeCodex
Route CodeFlowMu work to the right reviewer. Use when a technical manager needs to decide whether PM, QA, OPS, Architect, EVAL, or ADMIN should review a change, report, risk, or proposal.
Skill Claude CodeCodex
Analyze technical risks and dependencies for CodeFlowMu work. Use when a technical manager needs to identify blockers, dependency order, risk owners, runtime risk, missing upstream reports, or escalation paths.
Skill Claude CodeCodex
Organize CodeFlowMu UI screens, navigation, panels, filters, and operational workflows. Use when UI work needs information grouping, scan path, state/empty states, or safe navigation for task/report/lifecycle/log surfaces.
Skill Claude CodeCodex
Turn rough CodeFlowMu UI feedback into clear interface requirements. Use when designing Panel, desktop, task/report/log, lifecycle, or workflow UI needs user role, workflow, state semantics, constraints, and acceptance criteria.
Skill Claude CodeCodex
Accept or reject CodeFlowMu UI changes based on usability evidence. Use when QA, PM, or UI reviewer needs desktop/mobile checks, interaction results, visual overlap checks, and FCoP semantic correctness.
Skill Claude CodeCodex
Review CodeFlowMu UI for visual consistency, component fit, readable typography, responsive layout, and operational clarity. Use when checking Panel screens, modals, tables, filters, cards, buttons, or icons.
Skill Claude CodeCodex
Open selected webpages and extract relevant正文, structured facts, tables, and source URLs. Use when an agent has one or more URLs and needs faithful page content, dynamic-page rendering, table extraction, or citation-ready evidence without yet synthesizing a full research conclusion.