Interview the user to get unblocked after deep work, one question at a time. Triggered by 'ask me', 'I need your input', 'interview me to unblock this', or when only the human can answer.
Audit Claude Code hooks for safety and correctness. Use when reviewing hook health, checking hook safety, validating hooks, or after adding/modifying hooks.
Audit Popoto Redis models for gaps and naming drift. Use when reviewing data model health or the data layer, checking model integrity, validating Redis models, or scanning for model issues.
Audit tools/ for structure, tests, CLI quality, and docs. Use when checking tool health, validating tools, or after adding a tool. Also 'check the tools', 'which tools need work'.
Stage, launch, grade, and schedule a Claude Managed Agent (CMA) from a build-sheet. Triggered by 'build the agent', 'launch this agent', 'deploy a managed agent', or as /imagine-agent's handoff.
Reconstructs a day's work from git commit history in the current repo, groups it into time-blocked calendar events (merged by feature/goal, minimum 20 minutes each), and writes them to that repo's mapped Google Calendar — creating new events or updating existing overlapping ones so reruns stay idempotent. Replaces the…
Create/review/maintain Claude Code Routines (cloud scheduled agents, aka 'Cowork'). Triggered by 'set up a routine', 'schedule a cloud agent', 'cowork task', 'migrate to a routine'.
Pre-publish editorial gate for external-audience content: removes AI-writing tells and blocks hollow drafts. Use on documents, presentations, emails, posts, or web copy once a first draft exists. Triggered by 'de-slop this', 'editorial pass', 'humanize this', 'remove the AI tells', or invoked as a mandatory gate by…
Use when executing a plan document to ship a feature. Triggered by 'build this', 'execute the plan', 'implement the plan', or any request to run/ship a plan.
Send a spoken executive brief to a chat as a voice message. Triggered by 'send a voice debrief', 'speak this update', 'do-debrief', or any request to deliver an audio summary.
Audit a web UI against premium design criteria. Use when evaluating design quality, or on 'review this design', 'check the UI', 'audit this page', 'scan the interface', or a URL for design feedback.
Turn a moodboard into additive design-system edits. Triggered by 'apply this moodboard', 'tighten the design system', 'theme pass', 'design system pass', 'organize design files', or a moodboard URL.
Discover happy paths on a target site using BYOB MCP, producing trace JSON for deterministic test generation. Use when asked to discover paths, map site flows, or record browser traces for tests.
Deep audit of a named feature's integration health: orphan code, dead wiring, missing tests, config gaps. Use when asked to audit a feature's integration or wiring. User-invoked only.
Post a GitHub investigation issue for an unverified finding, gap, or anomaly. Use when an audit or research surfaces something suspicious, or an observed behavior is unexpected but unconfirmed.