agentleFS
Sign inStart free
BechsteinDigitalBechsteinDigital/claude-shiploop/autonomous-setup/SKILL.md

autonomous-setup

Skill4 starsChanged 4 months agoMITView on GitHub

Use when an approved product brief (project/BRIEF.md, status APPROVED) exists and the project should be set up without further user input — research, tech decisions, scaffold, backlog. Do not use while the brief is missing or DRAFT (→ project-onboarding).

1 other project uses a version of this.

Adopted by

Most recently taken up by

  • Other versions1 project uses a version of this.

    Word for word · 0

    None found.

    Edited · 0

    None found.

    Wrote their own version · 1

    1. go-steer/core-agent★ 9same name

    How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.

  • In this repository7 other files it gives its agents.

The file

We split this skill into the rules it gives an agent

1 rule. About 0 are common, 0 are distinctive, and 1 is very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈0

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 1

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Create project/PROFILE.md from _shared/templates/PROFILE.template.md.

    line 42

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Autonomous Setup

Principle

From here on, no questions to the user. Every open decision is made autonomously and logged in project/DECISIONS.md as an ADR. Only exceptions: the escalation criteria from the autonomy contract in project/BRIEF.md.

No AskUserQuestion tool (enforced): a reversible open decision → decide and log an ADR; an escalation criterion hit → stop with a decision memo. Never an interactive prompt.

Precondition

project/BRIEF.md exists with status APPROVED. Otherwise abort and request project-onboarding.

Phases

1. Research (timeboxed, parallel)

First read the global knowledge base (resolve: $SKILLS_KNOWLEDGE_DIR if set, otherwise the path in <skills-dir>/_shared/knowledge.path — written by install.sh; if neither resolves → skip) — documented learnings from earlier projects replace research and prevent repeat mistakes.

Then parallel read-only subagents in one invocation block, one per question:

  • Alternatives: What already exists? What does it not do well enough? (confirms/sharpens the core contract)
  • Tech options: 2–3 stack candidates fitting the must-have outcomes, target environment, and cost frame — with trade-offs, not with a winner
  • Risks: legal, technical, and cost traps of the domain

Results go to project/log/<date>-research-<topic>.md. Research findings that suggest new features go into project/IDEAS.md — never directly into scope.

2. Decide

Determine stack, architecture cut, and project structure: one ADR each in project/DECISIONS.md (decision, alternatives, rationale, reversal cost). On a tie, the decider is: least complexity for the must-have outcomes — not the most interesting technology.

3. Scaffold

  • Initialize the repo (if not present), skeleton per the ADRs, .gitignore, minimal README.
  • Create project/PROFILE.md from _shared/templates/PROFILE.template.md. Enter each command (build, test, lint, run) only after it ran successfully once.
  • A walking skeleton is enough: buildable, testable, startable. No feature implementation during setup.

4. Backlog

  • At least one WORK card per must-have outcome from the brief (project/backlog/, follow the template): P0 = critical path to MVP.
  • Cut cards small and with disjoint claim zones where possible — that enables the loop to parallelize.
  • Initialize project/STATE.md and project/IDEAS.md from the templates.

5. Finish

First commit (scaffold + project/ artifacts). Short setup report in project/log/. Then start autonomous-loop directly — don't wait for confirmation.

Red flags

  • "I'll quickly ask the user which language they prefer" → violation: decide, log, continue.
  • Research takes longer than the scaffold → timebox violated; decide with available knowledge.
  • Commands in the profile that never ran → the profile lies, the loop breaks later.
  • Setup already implements features → belongs in WORK cards and the loop.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

Check your setup, free