Agent
Mechanical engineer rung for well-specified execution: boilerplate, straightforward changes, simple edits, repetitive work. Implements via TDD when the task changes code, and commits changes before reporting. Pinned to Sonnet.
Agent
Mechanical engineer rung for well-specified execution: boilerplate, straightforward changes, simple edits, repetitive work. Implements via TDD when the task changes code, and commits changes before reporting. Pinned to Sonnet.
Agent
Deep-reasoning engineer rung for reasoning-heavy phases: architecture-adjacent implementation, complex or multi-system changes, debugging fixes, algorithm design. Implements via TDD and commits changes before reporting. Pinned to Opus so the orchestrator's frontier model is reserved for orchestration.
Agent
Executes marketing artifacts: campaigns, ads and creative, copy and copy-editing, cold/outbound email, landing-page content, SEO/AI-search artifacts, launches at execution level, customer/VOC research runs, and recurring marketing loops. Grounds every asset in real source material and cites the docs/marketing/…
Agent
Executes user-facing product design: UX, psychology, behavioural economics, network effects, growth — thought through holistically. Maintains the design system and product knowledge in docs/product/ (incl. PERSONAS.md). Grounds every decision in realistic, evidence-backed personas and returns Approve/Refine/Reject…
Agent
Verifies a change works by exercising the running app/feature from the outside (via the /verify skill) and reporting PASS/FAIL with evidence. Does not write code or fix failures.
Command
Write a handoff document so a fresh agent can continue this work after the conversation ends. Invoke when the user is about to /clear or /compact, when context is filling up, or when the user asks to hand off, checkpoint, or pause work for a later session. Argument (optional): what the next session will focus on.
Command
Audit how well a repository's codebase and practices align with the promode methodology, then produce a prioritised, actionable improvement plan. Runs via the promode:auditor agent, which fans out parallel assessors (one per dimension) and synthesises their findings. Use when the user wants to assess promode…
Hook
Runs when a session starts on startup, resume, clear and compact, running bash (20 commands). From mikekelly/promode.
Hook
Runs when the agent finishes a response, running bash. From mikekelly/promode.