agentleFS
BechsteinDigitalBechsteinDigital/claude-shiploop/project-onboarding/SKILL.md

project-onboarding

Skill4 starsChanged 4 months agoMITView on GitHub

Use when the user pitches a new project or product idea, wants to start a new project, or asks for an onboarding interview — including vague ideas ("I have an idea for an app"). The only interactive skill of the suite; everything after it runs autonomously.

13 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use code-review.

Other files people use alongside this one.

Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
  • GoogleSkill

    code-review

    Adversarial systems, concurrency, memory-model, and architecture code review protocol. Use this skill whenever asked to review code, PRs, diffs, or branches in Filament.

    Also in 3 projects that use this

The file

We split this skill into the rules it gives an agent

7 rules. 0 are common, 4 are distinctive, and 3 are very specific to this project.

  • 0

    Common

    Most of the 11 edited versions we read keep these.

  • 4

    Distinctive

    This author's own choices. Worth a look.

  • 3

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Only start the interview once the user has confirmed or corrected the reflection.

    line 19
  2. Very specific:

    Ask questions with AskUserQuestion (max.

    line 22
  3. Distinctive:

    After each round, briefly summarize what is now settled.

    line 31
  4. Distinctive:

    No leading questions that grow the scope ("Wouldn't you also like …?").

    line 34
  5. Very specific:

    Create project/BRIEF.md from _shared/templates/BRIEF.template.md (status: DRAFT), show the user the key points compactly and obtain …

    line 46
  6. Very specific:

    After approval, start autonomous-setup immediately.

    line 51
  7. Distinctive:

    Must-have outcomes that aren't phrased verifiably ("should look good")

    line 57

“Common” is measured: kept by at least 6 in 10 of the 11 edited versions of this file we could read. “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

Project Onboarding (Pitch Interview)

Goal

Extract all information from a pitch until autonomous work is possible — captured as an approved project/BRIEF.md. This is the last opportunity to ask the user questions; afterwards autonomous-setup takes over without follow-ups.

Process

1. Mirror the pitch

Reflect the idea back in 2–3 of your own sentences (core idea, assumed user, assumed problem). Only start the interview once the user has confirmed or corrected the reflection.

2. Interview in rounds

Ask questions with AskUserQuestion (max. 4 per round), one topic block per round, rounds build on each other. Question catalog: references/interview-guide.md — adapt round and wording to the answers there; don't work through it mechanically.

Order: A Core (problem, user, success) → B Scope (must-have outcomes, non-goals, smallest shippable version) → C Frame (time, cost, tech constraints, platform) → D Autonomy (escalation rules, extension budget, deployment limits).

Rules:

  • After each round, briefly summarize what is now settled.
  • "Don't know" → propose a default yourself and only have it confirmed — never drill deeper.
  • Technical detail questions only if the user is visibly technical; otherwise ask about effect, not technology.
  • No leading questions that grow the scope ("Wouldn't you also like …?"). The interview exists to sharpen the core, not to collect features.
  • Spontaneous feature ideas — whether from the user ("Oh, and X would be cool too!") or the interviewer: acknowledge briefly, note as a candidate for project/IDEAS.md and possibly as a non-goal, don't go deeper, return to the core. Only if the user insists it belongs to the core is it treated as a must-have-outcome candidate.

3. Check the Definition of Ready

Checklist in references/interview-guide.md#definition-of-ready. Every open item → a targeted follow-up round or a marked default. Don't start while any item is neither answered nor covered by a confirmed default.

4. Write the brief and get it approved

Create project/BRIEF.md from _shared/templates/BRIEF.template.md (status: DRAFT), show the user the key points compactly and obtain explicit approval. Only after approval: set status to APPROVED.

5. Handover

After approval, start autonomous-setup immediately. From here on, no more questions to the user — the autonomy contract in the brief governs the only exceptions.

Red flags

  • Interview starts without a confirmed reflection
  • More than ~4 rounds without an interim summary → user fatigues; propose defaults
  • Must-have outcomes that aren't phrased verifiably ("should look good")
  • Setup starts although the brief is still DRAFT

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.