UniPi

Long-running iterative development loops. Run arbitrarily-long tasks without diluting model attention. Triggers: ralph, ralph loop, iterative loop, long-running task, development loop.

Full compactor reference — tool parameters, anti-patterns, sandbox languages, context budget, workflows.

Diagnostics — validate config, session DB, runtimes, and troubleshoot compactor issues.

Context management — compact session, recall history, run code, search content.

Stats display — context savings, session metrics, compactions, sandbox and recall/search counters.

Helps you use the /unipi:btw side-conversation workflow effectively. Use when you want to think in parallel, ask side questions without interrupting ongoing work, or inject a side thread back into the main agent.

CocoIndex content indexing and search — use when you need to search indexed project content or trigger re-indexing

Active bug investigation — reproduce, diagnose, root-cause analysis. Produces debug report for /unipi:fix.

Collaborative discovery — explore problem space, evaluate approaches, write design spec. Use before any creative work.

Strategic planning — create implementation plan from brainstorm specs. Tasks, dependencies, acceptance criteria.

Review work — check task completion, run lint/build, mark reviewer remarks. Use after /unipi:work completes.

Execute plan — implement tasks, test, commit on done. Works in worktree or on main branch. Resumable.

Interactive decision-gating tool for structured user input. Use ask_user when you need user confirmation, preferences, or decisions before proceeding with high-impact or ambiguous choices.

Help user configure Pi notification settings — platforms (native, Gotify, Telegram), events, and per-event routing. Guide through setup or make changes directly.

Guide for using and configuring the Unipi updater — auto-update, changelog browser, and readme browser.

Web search, read, and summarize tools with provider-based backend

Research codebase — surface patterns, find prior art, prepare for brainstorm. Spawns subagents if available.

Diagnose and fix kanboard parser issues — validates all workflow documents, reports errors, suggests fixes.

Create MILESTONES.md from existing workflow docs — scan, propose, refine, write.

Sync MILESTONES.md with completed work — scan docs, diff checkboxes, auto-update.

Full pipeline — brainstorm → plan → work → review → merge. One command, end to end.

Create reusable chore — save repeatable tasks like deploy, publish, push to docs/chore/.

Execute a saved chore — run deploy, publish, push, or any repeatable task from docs/chore/.

Fix bugs using debug reports. Autocomplete for debug files. Records fixes in docs/fix/.

Fast bug fix without debug report. One-shot fix for clear, simple bugs. Records fixes in docs/fix/.

Fast single-task execution — one shot, done. Use for small tasks that don't need full brainstorm/plan/work flow.

Read-only research with bash access. Deep codebase investigation, documentation review, external research.

Passive codebase scan — find potential bugs, anti-patterns, security issues. Spawns subagents if available.

Cross-platform notification system for Pi. Use notify_user when you need to urgently alert the user about critical findings, errors, or completion of long-running tasks.

Guidance for using the @pi-unipi/utility extension commands and tools. Provides lifecycle management, diagnostics, cleanup, batch execution, environment inspection, and terminal utilities.

Merge worktree branches back to main. Gathers context from specs/plans, merges, cleans up.

MCP server management — discover, connect, and use Model Context Protocol tools

Persistent cross-session memory management. Store and retrieve user preferences, project decisions, code patterns, and conversation summaries across sessions. Use when you need to remember something important or recall past context.

List all unipi worktrees. Shows branch, path, and status.

Fast parallel codebase exploration

Parallel file writes with transparent locking

Consolidate — save learnings to memory, craft skills if reusable. Use at end of work session or to summarize current state.

Expert consultation — advisory with framework-based analysis. Read-only. Use when you need expert advice before brainstorming.

Generate documentation — README, API docs, guides. Use when you need to document code or features.

Create git worktree for parallel work. Agent-driven naming.