skillZs
★ LIVE SKILL TAGS ★
>>> LIVE SKILLS INDEX <<<
* OPEN SOURCE *
NO LOGIN, NO TRACKING
※ REAL INSTALL DATA ※
← back to all skills
thaitype/chief124 installs

chief-autopilot

Run chief-agent in full autopilot. Requires goals and contracts to exist. Chief creates TODO, delegates to builder, and repeats until milestone is done. Auto mode makes all decisions autonomously; safe mode stops on ambiguity. Use "/chief-autopilot" for auto or "/chief-autopilot safe" for safe mode.

How do I install this agent skill?

npx skills add https://github.com/thaitype/chief --skill chief-autopilot
view source ↗

Is this agent skill safe to install?

  • Gen Agent Trust Hubpass

    The skill implements an autonomous agent loop that processes project files to make decisions without human confirmation. This design presents an indirect prompt injection surface if an attacker can influence the goal or contract files.

  • Socketpass

    No alerts

  • Snykpass

    Risk: LOW · No issues

What does this agent skill do?

Run the chief-agent autonomously on the current milestone.

Arguments

  • No argument or auto → auto mode (default). Chief makes all decisions, never stops for human input.
  • safe → safe mode. Chief stops when ambiguity requires a design decision.

Prerequisite Check

Before doing anything:

  1. Identify the active milestone directory under .chief/.
  2. Check that _goal/ has at least one non-empty file.
  3. Check that _contract/ has at least one non-empty file.

If either is missing → STOP. Tell the user:

"Goals and contracts are required for autopilot. Run /chief-plan first."

Do NOT proceed.

Entry Confirmation

Present the current goals and contracts to the user in a brief summary (file names + 1-line description each).

Ask one question:

"Goals and contracts look correct? Proceed with full automation, or use /chief-plan to revise first?"

If the user says revise → stop. If the user confirms → proceed.

Execution Loop

1. Create or update TODO

Read existing _plan/_todo.md if present. Create the next batch of 3–5 tasks based on goals, contracts, and what's already done.

Write _plan/_todo.md. Do NOT wait for approval — this is autopilot.

2. Delegate to builder-agent

For each uncompleted task in _todo.md:

  • Delegate to builder-agent with:
    • The TODO entry
    • Milestone goals (_goal/)
    • Milestone contracts (_contract/)
    • Relevant global rules (.chief/_rules/)
    • Verification expectations
  • Wait for builder to complete
  • Mark task [x] in _todo.md when done

3. Handle blockers and ambiguity

Auto mode:

  • If builder reports a blocker or ambiguity → chief-agent picks the best option and continues.
  • Document the decision in the batch report (see below).
  • Builder-agent should escalate to chief-agent when possible before giving up.

Safe mode:

  • If builder reports a blocker or ambiguity → STOP and present the issue to the user with options.
  • Wait for user decision, then continue.

4. Repeat

After completing a batch, check if the milestone goals are fully met:

  • If more work remains → go back to step 1, create next batch.
  • If goals are met → write final batch report and stop.

Batch Report

After each batch (or when stopping), write a report to:

.chief/<milestone>/_report/autopilot-run-batch-<N>.md

Where <N> is the next available number (1, 2, 3...).

The report MUST contain these sections:

# Autopilot Run Batch <N>

## Mode
auto | safe

## Summary
What was accomplished in this batch.

## Tasks Completed
- task-1: ...
- task-2: ...

## Decisions Made (auto mode only)
For each ambiguity encountered:
- **Issue:** what was ambiguous
- **Options:** what choices existed
- **Chosen:** which option was picked
- **Reason:** why

## Backlog
Remaining work not yet done.

## User Action Needed
Items that require human decision or manual intervention.

Rules

  • NEVER start without goals and contracts existing.
  • NEVER skip the entry confirmation.
  • In auto mode, NEVER stop for human input — make decisions and document them.
  • In safe mode, ALWAYS stop on ambiguity — never guess.
  • Follow the rules hierarchy: AGENTS.md > .chief/_rules > milestone goals.
  • Builder-agent handles all implementation. Chief NEVER writes code.
  • Tester-agent is NOT used unless the user explicitly requests it.

Add the canonical catalog link to the repository README so users can inspect current installs and available audits. The publishing guide covers the complete discovery path.

<a href="https://skillzs.dev/skills/thaitype/chief/chief-autopilot">View chief-autopilot on skillZs</a>