Explore skills

Generates physical world foundations: biomes, planetary systems, climate, and ecological zones with embedded narrative tension. Use when creating a new world, planet, biome, or ecosystem, or when user says "new world", "new planet", "new biome", "build a world", "create a setting", or "environmental design". Does NOT write final files — use /writing-worldbuilding for that.

Search indexed documents (PDF, Word, Excel, PPT, TXT) using BM25 keyword search. Use for EDA tools, company docs, bug reports, and technical documentation.

Read-and-revise pass for any drafted post. Applies writing-style.md checklist, tightens prose, kills filler. Run after drafting and before ai-slop-check.

Post-launch breakdown (~800-1500 words). What you shipped, the numbers (good or bad), what worked, what didn't, what you'd change next time. Use after every product/launch/release where you have real numbers to share — including (especially) when the numbers are embarrassing.

Vision/strategy piece (~1500-3000 words). Rare format — once or twice a year. Names what you're building, why it matters, and how you'll get there. Use to plant a flag, reset direction, or invite people into a longer-term mission.

On-demand write protection for approved files during final assembly, implemented via Claude Code PreToolUse hooks — a rare working example of hook-based file guards. Use when the user enters a final-assembly phase, wants to freeze approved work, or says "lock scenes", "protect these files", "freeze approved", "/lockscenes", "add a write guard", or asks how to temporarily block edits during a criti

Batch-generate 2x2 contact sheet grids via the Nanobanana MCP (Nano Banana / NB2 / Gemini Image) for AI media pipelines. Handles prompt cleaning, up to 3 reference images per call, parallel batch orchestration, and generation logging so prior runs inform new ones. Use whenever the user asks to "generate frames", "run NB2 batch", "regen scenes", "generate contact sheets", "make grid variants", "bat

Miki's personal pre-self-review loop. Iteratively runs /review on the current PR and addresses obvious issues until only minor nits remain or 3 iterations have run. Surfaces judgment calls back to the user, makes atomic commits per project convention, pushes once at the end. Invoke when the user says "/miki-review-loop", "loop the review", "iterate on the review", or similar after a PR has been cr

Miki's session wrap-up check. Surveys the current conversation and working tree for anything outstanding before ending the session — open todos, uncommitted changes, unpushed commits, open PRs that haven't merged, background agents still running, follow-ups the conversation flagged but never closed, and stale references noticed but not fixed. Produces a short punch list, or confirms there's nothin

Analyze an implementation plan (defaults to .flow/specs/), discover relevant skills from the ecosystem, install them, and rewrite the plan to leverage installed skills. Use when user has a build plan, flow-next epic spec, or feature request and wants to equip the agent with the best available skills before building. Triggers on 'select skills for', 'equip for plan', 'find skills for this plan', 's

Detection pass for AI-generated patterns and clichés. Final check before publishing. Returns a list of detected issues with severity and fix suggestions.

Contrarian opinion piece (~800-1500 words). Stakes a position against conventional wisdom in your niche. Use when you have a real fight to pick — a widely-held belief in your space that you genuinely disagree with and can defend with evidence.