Agent
Engineering Manager — designs HOW to build through engineering definitions, architecture, plans, and technical contracts. Owns engineering-level goals, prompt architecture, and QA. Use when defining mechanisms, designing architecture, creating plans, defining data contracts, auditing prompts, or evaluating technical…
Agent
Product Manager — defines WHAT to build through product definitions, specs, and user research. Owns product-level goals. Use when defining features, writing specs, or processing feedback.
Agent
Single-resource debate advocate. Champions ONE resource (time, cost, scope, ux, tech-debt, risk, scale, simplicity) in the thinking-partner adversarial debate. Dispatched per advocate per round; never balanced — always opinionated.
Agent
Direction judgment worker — assesses strategic artifact coherence. Reads why, vision, stage, and goals to verify internal consistency. Writes structured assessment for the direction orchestrator.
Agent
Planning judgment worker — assesses alignment between product definitions, engineering definitions, and specs. Writes structured assessment for the planning orchestrator.
Agent
Shipping judgment worker — verifies implementation quality across security, functionality, and code quality dimensions. Can dispatch QA sub-skills (ux-audit, semantic-qa, qa-visual). Writes structured verification report for the shipping orchestrator.
Agent
Cognitive thinking partner for working through decisions, trade-offs, and unknowns. Use when exploring options, challenging assumptions, or thinking through complex problems before acting.
Agent
Strategic visionary — sets project direction, stage, quality constraints, and business goals. Owns the WHY. Use when setting vision, defining stage, or evaluating business metrics.
Skill Claude CodeCodex
Land one architecture decision atomically: full entry to architecture-archive.json, hard-capped lean entry to architecture.json. Handles new decisions, supersession, amendment. Triggers: 'record this decision', 'log an ADR', 'we decided', 'supersede ADR', 'that replaces the earlier decision'.
Skill Claude CodeCodex
Maintain a code-derived 'current belief of the system architecture' map — key applications, datastores, contracts, and integration points. Triggers: 'map architecture', 'architecture map', 'what does the system look like', 'refresh architecture map'.
Skill Claude CodeCodex
Update CLAUDE.md with learnings, preferences, or project context. Use when user says "update claude.md", "add to claude.md", or wants to persist something.
Skill Claude CodeCodex
Audit a codebase for dead code, orphaned modules, unused dependencies, and unwired seams. Quick tier runs knip-class checks ephemerally (no repo mutation); deep and exhaustive tiers add intent↔code reconciliation. Portable to any repo, including ones with no tooling installed.
Skill Claude CodeCodex
Generate project index for faster codebase navigation. Triggers: 'index codebase', 'create index', 'map project'.
Skill Claude CodeCodex
Use when user wants to visualize or present shipkit content as HTML. Triggers: 'create presentation', 'visualize this', 'HTML report'.
Skill Claude CodeCodex
Scaffold a tiered design system — principles, tokens, aesthetic direction from project context. Triggers: 'design system', 'design tokens', 'brand guidelines', 'visual direction', 'design principles'.
Skill Claude CodeCodex
Define a project's WHY/WHAT foundation autonomously — the DIRECTION entry point. Drives the definitional steps to a coherent bar via the engine (autonomous-propose): grounds every field in a cited signal, asks only the few high-leverage decisions. Triggers: 'start a project', 'define direction', 'kick off', 'set up…
Skill Claude CodeCodex
Define the technical approach — mechanisms, design decisions, stack direction, and component structure. Triggers: 'engineering approach', 'how to build this', 'technical design', 'mechanisms'.
Skill Claude CodeCodex
Derive technical performance criteria from the engineering blueprint. Writes goals/engineering.json with response times, reliability, coverage thresholds. Evaluate mode checks actuals against targets.
Skill Claude CodeCodex
Process user feedback into investigated bug specs with root cause analysis. Triggers: 'triage feedback', 'process bug reports', 'user testing feedback'.
Skill Claude CodeCodex
Discover and install MCP servers for enhanced Claude capabilities.
Skill Claude CodeCodex
Discover and install skills and plugins from the Claude Code ecosystem.
Skill Claude CodeCodex
Capture current metric values for goal evaluation. Triggers: 'log metrics', 'update metrics', 'record measurements'.
Skill Claude CodeCodex
Core automation engine. Drives a set of steps to a confirmed ground-truth bar via a main-session delegate → reconcile → re-dispatch loop. Phase skills (build, review, direction) call it; can also run directly. Two modes: autonomous (run to done unattended) and steered (coordinate, surface to user). Triggers: invoked…
Skill Claude CodeCodex
Use when a spec exists and user wants implementation steps. Triggers: 'how to implement', 'create plan', 'plan this', 'what are the steps'.