ai-outfitter

Welcome new DeepWork users on Pi and introduce native Pi workflows

Record a workflow by doing it, then turn it into a DeepWork job

Start or continue DeepWork workflows using native Pi tools

Set up DeepWork Reviews using .deepreview config files

Start structured planning that produces an executable DeepWork plan

Reference documentation for DeepWork Reviews

Create and manage DeepSchemas with validation and review generation

Run DeepWork Reviews on the current branch using .deepreview rules

Transcript-driven video post-production — set up the toolchain, transcribe with whisper.cpp, and cut/speed/export with ffmpeg. Load the matching reference for detailed commands.

Use when refining or reviewing ideas, documents, and communications for clarity and logical structure.

Triage one newly opened issue on the community-profiles catalog: classify it as feat (new agent, skill, or prompting change) or fix, apply that one label, and comment with a plan and example pseudo-code following Outfitter best practices.

Explain Outfitter and help users compose, inspect, and maintain .agents resources, agents and their loadouts, personas, skills, and settings. Use when a user asks about Outfitter itself — the .agents protocol, agent profiles and personas, catalogs, skills, state persistence, launch configuration — or asks to set up, change, debug, or migrate an Outfitter-managed launch.

Triage one newly opened issue on the Outfitter CLI repository: classify it as bug, enhancement, or question, apply that one label, and comment with an assessment and a concrete plan grounded in the repository's docs.

Set up, review, or simplify agentic GitHub workflows built with ai-outfitter/actions. Use when adding GitHub-triggered agents, reducing duplicate profiles or jobs, defining structured trigger context, or deciding how profiles and skills should divide responsibility.

Run the bundled data analyst demo workflows from R2-backed DeepWork demo bundles

Handle Slack app mentions from Channels wakes using channel_read and channel_respond. Use when a wake contains an opaque Slack locator.

Generate ai-outfitter organization reports from GitHub data. Use for any CI run whose trigger_context carries a report_kind (e.g. weekly-kpi) — the skill routes each kind to its report flow, collection scripts, and template.