PluginBench
Skill
Review
Audit score 70

pi-delegate

amelnagdy/delegate-skills

Delegate coding tasks to Pi agent CLI, review diffs, and land verified changes yourself.

What is pi-delegate?

Orchestrate bounded coding work by dispatching tasks to the Pi coding agent as a background implementer, then reviewing its output and committing verified changes. Use this when the user explicitly asks to delegate implementation work to Pi or run a queue of tasks through it while staying in control of review and landing.

  • Dispatch a coding task brief to Pi via the bundled relay script
  • Capture Pi's JSON event stream and write results to result.json
  • Review Pi's diff against the brief before landing changes
  • Resume interrupted sessions with delta briefs for rework
  • Support read-only runs for diagnosis without modifications
  • Enforce human orchestrator control: Pi implements, you commit

How to install pi-delegate

npx skills add https://github.com/amelnagdy/delegate-skills --skill pi-delegate
Prerequisites
  • Install Pi with `npm install -g @earendil-works/pi-coding-agent`
  • Authenticate Pi via `/login` or API-key environment variable / auth file
  • Confirm `pi --version` succeeds
  • Work in or point `--cd` at a git repository
Claude Code
Cursor
Windsurf
Cline

How to use pi-delegate

  1. 1.Write a clear brief describing the goal, current state, changes needed, and what to leave untouched; include the project's actual gates and report contract
  2. 2.Run the relay: `node <skill-dir>/scripts/relay.mjs --brief brief.txt --cd /path/to/repo` (optionally add `--model`, `--provider`, `--read-only`, `--approve`, `--timeout`, or `--resume-last`)
  3. 3.Wait for the relay to complete; it blocks until Pi finishes and writes `result.json`
  4. 4.Review the diff in `result.json` against your brief: check `touchedFiles`, re-run project gates, read the changes, and inspect Pi's final message as a claim not a guarantee
  5. 5.Commit only after gates pass and the diff holds; if rework is needed, send a delta brief with `--resume-last` and review again

Use cases

Good for
  • Delegate a feature implementation to Pi while you review and land the changes
  • Run a queue of coding tasks through Pi overnight, then review and commit each one
  • Ask Pi to refactor a module, review the diff, and land it after gates pass
  • Resume an incomplete implementation session with additional context or corrections
  • Run a read-only diagnostic pass to understand what Pi would change without committing
Who it's for
  • Developers who want to delegate implementation work but retain review and commit control
  • Teams running batches of coding tasks through an agent while staying the gatekeeper
  • Engineers who need to verify agent output against project gates before landing

pi-delegate FAQ

When should I use pi-delegate vs. writing code inline?

Use pi-delegate for bounded tasks large enough that delegation overhead is worth it and when the user explicitly asks to delegate to Pi. For small inline changes or when the user wants code written directly without delegation, skip it.

What if Pi's output doesn't match the brief or gates fail?

Send a delta brief with `--resume-last` or `--session <id>` to resume the session with corrections or additional context. Review the new diff the same way before committing.

Is Pi sandboxed or permission-restricted?

No. Pi has no sandbox and no permission modes. Use `--read-only` to restrict tools to read, grep, find, and ls for diagnosis. Use `--no-approve` (default) to keep project `.pi` settings untrusted. Always inspect `touchedFiles` and the diff after every run.

Can I run multiple tasks through Pi in sequence?

Yes. Use the multi-task-queue pattern: write briefs for each task, dispatch them sequentially (or background them), review and land each one after gates pass. See references/multi-task-queues.md for constraint carry-forward and coherence checks.

What if the relay fails or Pi is unavailable?

A pre-run usage error exits with code 2 and writes no result. A missing `pi` exits 127 and writes `status: "pi_unavailable"` to result.json. Check `pi --version` and authentication before retrying.

Full instructions (SKILL.md)

Source of truth, from amelnagdy/delegate-skills.


name: pi-delegate description: >- Delegate a coding task to the Pi coding agent CLI (pi) as a background implementer, then review its diff and land it yourself. Use this whenever the user wants to delegate implementation work to Pi - phrasings like "have Pi implement X", "delegate this to pi", "run it through Pi", or "use pi to implement/fix/refactor" - or wants to run a queue of coding tasks through Pi while staying the reviewer. DO NOT USE for tasks small enough to do inline, or when the user wants the code written directly without delegating. license: MIT metadata: version: 0.5.0

Pi Delegate

You are the orchestrator. Delegate a bounded coding task to a separate implementer - the Pi coding agent CLI - then review what it produced and land it yourself. You write the brief and own the judgment; the implementer makes changes in its own session; you verify and commit.

The loop needs only a shell command and file access, so any comparable orchestrator can drive it.

When NOT to use this

  • The task is small enough to do inline; delegation overhead is not worth it.
  • The pi CLI is not installed or authenticated.
  • You need a sandboxed implementer. Pi has no sandbox and no permission modes; --read-only restricts the tool surface, but a write-capable run executes without prompts.

Prerequisites (check once)

  1. Install pi with npm install -g @earendil-works/pi-coding-agent.
  2. Authenticate: /login inside pi for a subscription provider, or an API-key environment variable / pi's auth file for an API-key provider.
  3. Confirm pi --version succeeds.
  4. Work in, or point --cd at, the target git repository.

Choose the model (optional)

Omit --model to use pi's configured default. To pick another, choose from pi --list-models and pass an explicit id or pattern like <provider>/<model-id> or sonnet:high. The relay accepts letters, digits, and . _ : / - only (the value reaches a shell on Windows), so glob patterns with * are not forwarded.

The loop

Run these five steps per task. Steps 1, 4, and 5 require judgment; 2 and 3 are mechanical.

1. Write the brief

Pi sees only the text you send plus what it can inspect in the workspace - no chat history or shared context. Include the goal, current state, what to change, what to leave untouched, the project's actual gates, and a report contract. Tell pi not to commit. Keep one task per brief. Pi auto-loads AGENTS.md/CLAUDE.md context files from the workspace and its parents, so repo instructions reach it without inlining. See references/writing-the-brief.md.

2. Dispatch

Use the bundled relay. It pipes the brief to pi --mode json on stdin, captures the JSON event stream, and writes result.json. (<skill-dir> is the installed folder containing this SKILL.md.)

node "<skill-dir>/scripts/relay.mjs" --brief brief.txt --cd /path/to/repo
# choose a model:                          add --model <id from pi --list-models>
# choose a provider:                       add --provider <name>
# read-only run (review/diagnosis):        add --read-only
# trust project .pi resources:             add --approve
# resume the most recent session:          add --resume-last  (delta brief only)
# resume a specific session:               add --session <id> (delta brief only)
# hard time limit (watchdog):              add --timeout 2h  (the 30m default suits short runs; implementation briefs routinely need 1-2h)
# see all options:                         node .../relay.mjs --help

The child process's cwd pins the workspace. The relay writes artifacts under the system temp dir by default and never commits. See references/dispatch-and-poll.md.

3. Wait for completion

The relay blocks until pi finishes. Run it with the orchestrator's background-command facility, or background it in the shell and poll for result.json. A pre-run usage error exits 2 and writes no result; a missing pi exits 127 and writes status: "pi_unavailable".

Trust process state and the working tree over a progress display. Completion means the process exited and result.json exists. Pi's full report is the finalMessage field in result.json (also printed in full on stdout between the report markers).

4. Review - do not trust the self-report

Treat pi's final message and gate claims as claims:

  • Re-run the project's gates yourself.
  • Read the diff against the brief, starting with touchedFiles.
  • Run relevant guard skills if installed.
  • Round-trip migrations and grep for dangling references after removals or renames.

See references/review-and-land.md.

5. Land it

The implementer edits the working tree; the orchestrator commits. Commit only after the gates pass and the diff holds. If rework is needed, send a delta brief with --resume-last or --session <id>, then review again.

Autonomy and permissions

Pi has no sandbox and no permission modes. A default headless run reads, writes, edits, and executes shell commands with no prompts - the controls are:

  1. --read-only restricts pi's callable tools to --tools read,grep,find,ls across built-in, extension, and custom tools. Installed extension code still runs with the user's host permissions.
  2. The relay passes --no-approve by default, so project .pi settings, extensions, and skills stay untrusted. --approve is the explicit opt-in for a repository the user trusts.
  3. touchedFiles and the diff are the record of what changed. Inspect them after every run.

Authorization model

Delegation is something the human opts into. Once they have ("run this queue", "proceed"), committing verified, gate-passing work is the agreed contract. Two limits remain: surface, don't absorb (report pi's design decisions, defensible-but-unasked turns, and non-blocking nitpicks) and stop for scope changes (if correct completion needs going beyond the brief, ask instead of expanding the mandate). See references/review-and-land.md.

References