Strict engineering discipline and agentic orchestration for the pi coding agent. Multi-agent workflows, executable plans, and milestone-level decomposition. pi μ½λ© μμ΄μ νΈλ₯Ό μν μ격ν μμ§λμ΄λ§ λμ¬μ΄νλ¦°κ³Ό μμ΄μ νΈ μ€μΌμ€νΈλ μ΄μ . λ€μ€ μμ΄μ νΈ μν¬νλ‘, μ€ν κ°λ₯ν κ³ν, λ§μΌμ€ν€ λΆν΄.
Every line is on GitHub. No hidden prompts, no secret system instructions, no obfuscated behavior. Tools, event hooks, skill injections, and agent prompts are all plain TypeScript and Markdown.
λͺ¨λ μ½λκ° GitHubμ 곡κ°λμ΄ μμ΅λλ€. μ¨κ²¨μ§ ν둬ννΈ, λΉλ° μμ€ν μ§μμ΄, λλ νλ λμ μμ. λꡬ, μ΄λ²€νΈ ν , μ€ν¬ μ£Όμ , μμ΄μ νΈ ν둬ννΈκ° λͺ¨λ μΌλ° TypeScriptμ MarkdownμΌλ‘ μμ±λμ΄ μμ΅λλ€.
The footer displays real-time metrics every session: prompt cache hit rate, context usage bar, active tools, branch, and model. See exactly how your context is being utilized.
νΈν°μ λ§€ μΈμ μ€μκ° λ©νΈλ¦μ΄ νμλ©λλ€: ν둬ννΈ μΊμ μ μ€λ₯ , 컨ν μ€νΈ μ¬μ©λ λ°, νμ± λꡬ, λΈλμΉ, λͺ¨λΈ. 컨ν μ€νΈκ° μ΄λ»κ² μ¬μ©λλμ§ μ νν νμΈνμΈμ.
No magic. Read the source and see exactly what the agent does. Skill rules are plain Markdown. Agent prompts are plain text. Every event hook is auditable.
λ§λ² μμ. μμ€λ₯Ό μ½κ³ μμ΄μ νΈκ° μ νν 무μμ νλμ§ νμΈνμΈμ. μ€ν¬ κ·μΉμ μΌλ° Markdown, μμ΄μ νΈ ν둬ννΈλ μΌλ° ν μ€νΈ, λͺ¨λ μ΄λ²€νΈ ν μ΄ κ°μ¬ κ°λ₯ν©λλ€.
We prefer straightforward solutions over elaborate orchestration. Fewer moving parts, fewer failure modes. See our contributing guidelines.
볡μ‘ν μ€μΌμ€νΈλ μ΄μ λ³΄λ€ λ¨μν ν΄κ²°μ± μ μ νΈν©λλ€. μμ§μ΄λ λΆλΆμ΄ μ μμλ‘ μ€ν¨λ μ μ΅λλ€. κΈ°μ¬ κ°μ΄λλΌμΈμ μ°Έμ‘°νμΈμ.
The extension provides a three-phase workflow: clarify β plan β execute. Each phase is driven by a slash command.
μ΅μ€ν μ μ 3λ¨κ³ μν¬νλ‘λ₯Ό μ 곡ν©λλ€: λͺ νν β κ³ν β μ€ν. κ° λ¨κ³λ μ¬λμ λͺ λ ΉμΌλ‘ ꡬλλ©λλ€.
/setup
β Configure Settings
β μ€μ ꡬμ±
Run once after installation to apply recommended settings. Sets quietStartup: true so the extension's custom ROACH PI banner replaces the default startup listing.
μ€μΉ ν ν λ² μ€ννμ¬ κΆμ₯ μ€μ μ μ μ©ν©λλ€. μ΅μ€ν
μ
μ 컀μ€ν
ROACH PI λ°°λκ° κΈ°λ³Έ μμ λͺ©λ‘μ λ체νλλ‘ quietStartup: trueλ₯Ό μ€μ ν©λλ€.
/clarify
β Resolve Ambiguity
β λͺ¨νΈμ± ν΄κ²°
The agent asks dynamic, context-aware questions one at a time while exploring the codebase in parallel. Ends with a structured Context Brief that defines scope, constraints, and success criteria.
μμ΄μ νΈκ° μ½λλ² μ΄μ€λ₯Ό λ³λ ¬λ‘ νμνλ©΄μ λμ μ΄κ³ λ¬Έλ§₯μ λ§λ μ§λ¬Έμ νλμ© λ¬»μ΅λλ€. λ²μ, μ μ½, μ±κ³΅ κΈ°μ€μ μ μνλ ꡬ쑰νλ 컨ν μ€νΈ λΈλ¦¬νλ‘ λλ©λλ€.
/plan
β Create an Executable Plan
β μ€ν κ°λ₯ν κ³ν μμ±
Transforms the Context Brief into an implementation plan with exact file paths, code blocks, commands, and expected outputs. No placeholders β every step is immediately executable by a worker agent.
컨ν μ€νΈ λΈλ¦¬νλ₯Ό μ νν νμΌ κ²½λ‘, μ½λ λΈλ‘, λͺ λ Ήμ΄, μμ μΆλ ₯μ΄ ν¬ν¨λ ꡬν κ³νμΌλ‘ λ³νν©λλ€. νλ μ΄μ€νλ μμ β λͺ¨λ λ¨κ³κ° μ컀 μμ΄μ νΈκ° μ¦μ μ€νν μ μμ΅λλ€.
/ultraplan
β Milestone Decomposition
β λ§μΌμ€ν€ λΆν΄
For complex multi-day tasks. Dispatches 5 independent reviewer agents in parallel (Feasibility, Architecture, Risk, Dependency, User Value), then synthesizes findings into an optimized milestone DAG.
볡μ‘ν λ©°μΉ μ§λ¦¬ μμ μ μν΄. 5κ°μ λ 립μ μΈ λ¦¬λ·°μ΄ μμ΄μ νΈλ₯Ό λ³λ ¬λ‘ λμ€ν¨μΉ(νλΉμ±, μν€ν μ², 리μ€ν¬, μμ‘΄μ±, μ¬μ©μ κ°μΉ)ν ν, μ΅μ νλ λ§μΌμ€ν€ DAGλ‘ μ’ ν©ν©λλ€.
/loop
β Session Loop Scheduler
β μΈμ
루ν μ€μΌμ€λ¬
Schedule recurring prompts at fixed intervals. Cron-style timing β fires on schedule regardless of execution state. Supports 5s, 10m, 2h, 1d formats. Up to 100 concurrent jobs with per-job error isolation and automatic cleanup on session shutdown.
κ³ μ κ°κ²©μΌλ‘ λ°λ³΅ ν둬ννΈλ₯Ό μ€μΌμ€λ§ν©λλ€. Cron μ€νμΌ νμ΄λ° β μ€ν μνμ κ΄κ³μμ΄ μ€μΌμ€λλ‘ μ€νλ©λλ€. 5s, 10m, 2h, 1d νμμ μ§μν©λλ€. μ΅λ 100κ° λμ μμ
, μμ
λ³ μλ¬ κ²©λ¦¬, μΈμ
μ’
λ£ μ μλ μ 리.
π‘ Tip: The ask_user_question tool is always available β the agent will ask clarifying questions autonomously whenever it detects ambiguity, even outside /clarify mode.
π‘ ν: ask_user_question λꡬλ νμ μ¬μ© κ°λ₯ν©λλ€ β /clarify λͺ¨λ μΈμμλ λͺ¨νΈμ±μ κ°μ§νλ©΄ μλμΌλ‘ μ§λ¬Έν©λλ€.
12 bundled agents for exploration, execution, planning, and review. Delegate tasks via single, parallel (max 8 tasks, 4 concurrent), or chain modes with cycle detection and depth guards.
νμ, μ€ν, κ³ν, 리뷰λ₯Ό μν 12κ° λ²λ€ μμ΄μ νΈ. μ¬μ΄ν΄ κ°μ§ λ° κΉμ΄ κ°λμ ν¨κ» λ¨μΌ, λ³λ ¬(μ΅λ 8κ° μμ , 4κ° λμ), μ²΄μΈ λͺ¨λλ‘ μμ μ μμνμΈμ.
Karpathy rules auto-injected into code-writing agents. Automatic slop-cleaner spawns after successful runs to remove LLM-specific patterns while preserving behavior.
Karpathy κ·μΉμ΄ μ½λ μμ± μμ΄μ νΈμ μλ μ£Όμ λ©λλ€. μ±κ³΅μ μΈ μ€ν ν λμμ μ μ§νλ©΄μ LLM νΉμ ν¨ν΄μ μ κ±°νλ μλ μ¬λ‘ ν΄λ¦¬λκ° μ€νλ©λλ€.
plan-compliance β plan-worker β plan-validator pipeline with information barriers. Validators never see execution context β they judge solely from the plan document and codebase.
μ 보 μ₯λ²½μ΄ μλ plan-compliance β plan-worker β plan-validator νμ΄νλΌμΈ. κ²μ¦μλ μ€ν 컨ν μ€νΈλ₯Ό λ³΄μ§ μμ΅λλ€ β κ³ν λ¬Έμμ μ½λλ² μ΄μ€λ§μΌλ‘ νλ¨ν©λλ€.
Custom ROACH PI ASCII banner, branded footer with directory, branch, model, context bar, cache hit rate, and active tools display.
λλ ν 리, λΈλμΉ, λͺ¨λΈ, 컨ν μ€νΈ λ°, μΊμ μ μ€λ₯ , νμ± λꡬ νμκ° μλ 컀μ€ν ROACH PI ASCII λ°°λ λ° λΈλλλ νΈν°.
Microcompaction truncates old tool results. Phase-aware summarization preserves workflow state across compaction. Phase and goal document survive context limits.
λ§μ΄ν¬λ‘μ»΄ν©μ μ΄ μ€λλ λꡬ κ²°κ³Όλ₯Ό μλ¦ λλ€. λ¨κ³ μΈμ μμ½μ΄ μ»΄ν©μ μ λ°μ κ±Έμ³ μν¬νλ‘ μνλ₯Ό 보쑴ν©λλ€. λ¨κ³ λ° λͺ©ν λ¬Έμκ° μ»¨ν μ€νΈ νκ³λ₯Ό μ΄κ³Όν©λλ€.
Schedule recurring prompts with /loop 5m check status. Cron-style fixed intervals, per-job error isolation, AbortController cancellation, timeout protection, and automatic cleanup on session shutdown. Up to 100 concurrent jobs.
/loop 5m check statusλ‘ λ°λ³΅ ν둬ννΈλ₯Ό μ€μΌμ€λ§ν©λλ€. Cron μ€νμΌ κ³ μ κ°κ²©, μμ
λ³ μλ¬ κ²©λ¦¬, AbortController μ·¨μ, νμμμ 보νΈ, μΈμ
μ’
λ£ μ μλ μ 리. μ΅λ 100κ° λμ μμ
.
Bundled LLM skill rulesets: clarification, plan crafting, run plan, review work, simplify, systematic debugging, Karpathy discipline, Rob Pike optimization, and more.
λ²λ€ LLM μ€ν¬ λ£°μ : λͺ νν, κ³ν μμ±, κ³ν μ€ν, μμ 리뷰, λ¨μν, 체κ³μ λλ²κΉ , Karpathy λμ¬μ΄νλ¦°, Rob Pike μ΅μ ν λ±.
pi is a terminal-based AI coding agent that helps you read, write, and debug code directly from your terminal. It supports extensions, custom tools, themes, and multiple AI providers. GitHub β
piλ ν°λ―Έλμμ μ§μ μ½λλ₯Ό μ½κ³ , μ°κ³ , λλ²κΉ νλ ν°λ―Έλ κΈ°λ° AI μ½λ© μμ΄μ νΈμ λλ€. μ΅μ€ν μ , 컀μ€ν λꡬ, ν λ§ λ° μ¬λ¬ AI νλ‘λ°μ΄λλ₯Ό μ§μν©λλ€. GitHub β
No. All engineering discipline skills are bundled with this extension. Just install the extension and run /setup β everything is ready out of the box.
μλμ. λͺ¨λ μμ§λμ΄λ§ λμ¬μ΄νλ¦° μ€ν¬μ΄ μ΄ μ΅μ€ν
μ
μ λ²λ€λ‘ ν¬ν¨λμ΄ μμ΅λλ€. μ΅μ€ν
μ
μ μ€μΉνκ³ /setupμ μ€ννμΈμ β λͺ¨λ κ²μ΄ μ¦μ μ¬μ© κ°λ₯ν©λλ€.
Single: One-off tasks sent to a specific agent.
Parallel: Dispatch multiple agents at once (max 8 tasks, 4 concurrent). Each runs independently.
Chain: Sequential pipeline where each step receives the previous step's output via {previous} placeholder.
All modes include cycle detection, depth limits (max 3), and concurrency control.
λ¨μΌ: νΉμ μμ΄μ νΈμ 보λ΄λ μΌνμ± μμ
.
λ³λ ¬: μ¬λ¬ μμ΄μ νΈλ₯Ό ν λ²μ λμ€ν¨μΉ(μ΅λ 8κ° μμ
, 4κ° λμ). κ°κ° λ
립μ μΌλ‘ μ€ν.
체μΈ: κ° λ¨κ³κ° μ΄μ λ¨κ³μ μΆλ ₯μ {previous} νλ μ΄μ€νλλ‘ λ°λ μμ°¨ νμ΄νλΌμΈ.
λͺ¨λ λͺ¨λμ μ¬μ΄ν΄ κ°μ§, κΉμ΄ μ ν(μ΅λ 3), λμμ± μ μ΄κ° ν¬ν¨λ©λλ€.
The slop-cleaner automatically runs after successful worker or plan-worker executions. It detects and removes LLM-specific code patterns (unnecessary comments, over-verbose logging, redundant type assertions, etc.) while preserving the original behavior.
μ¬λ‘ ν΄λ¦¬λλ worker λλ plan-worker μ€ν μ±κ³΅ ν μλμΌλ‘ μ€νλ©λλ€. μλ λμμ μ μ§νλ©΄μ LLM νΉμ μ½λ ν¨ν΄(λΆνμν μ£Όμ, κ³Όλν λ‘κΉ
, μ€λ³΅ νμ
λ¨μΈ λ±)μ κ°μ§νκ³ μ κ±°ν©λλ€.
Absolutely. The three slash commands are independent. Use /clarify + /plan for single-session tasks. Use /ultraplan only when you need milestone-level decomposition for complex, multi-day projects.
λ¬Όλ‘ μ
λλ€. μΈ κ°μ μ¬λμ λͺ
λ Ήμ λ
립μ μ
λλ€. λ¨μΌ μΈμ
μμ
μλ /clarify + /planμ μ¬μ©νμΈμ. 볡μ‘ν λ©°μΉ μ§λ¦¬ νλ‘μ νΈμ λ§μΌμ€ν€ μμ€ λΆν΄κ° νμν λλ§ /ultraplanμ μ¬μ©νμΈμ.