agentleFS
agoopintsaiagoopintsai/shipright/skills/setup/SKILL.md

setup

Skill2 starsChanged 16 days agoApache-2.0View on GitHub

Set up Shipright for this project - create or connect the brain repo, choose red zones, write shipright.yaml, and show the gate once. Run it when Shipright says it is not configured or its config is broken.

191 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use add-voice-transcription, add-discord and 10 more.

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.
  • juancruzmunozalbeloSkill

    add-voice-transcription

    Add voice message transcription to NanoClaw using OpenAI's Whisper API. Automatically transcribes WhatsApp voice notes so the agent can read and respond to them.

    Also in 5 projects that use this

  • MicrosoftSkill

    add-discord

    Add Discord bot channel integration via Chat SDK.

    Also in 11 projects that use this

  • MicrosoftSkill

    add-slack

    Add Slack channel integration via Chat SDK.

    Also in 11 projects that use this

  • juancruzmunozalbeloSkill

    add-telegram-swarm

    Add Agent Swarm (Teams) support to Telegram. Each subagent gets its own bot identity in the group. Requires Telegram channel to be set up first (use /add-telegram). Triggers on "agent swarm", "agent teams telegram", "telegram swarm", "bot pool".

    Also in 7 projects that use this

  • +8 more from microsoft

The file

Show its rules and text

We split this skill into the rules it gives an agent

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

  • 0

    Common

    Most of the 171 edited versions we read keep these.

  • 7

    Distinctive

    This author's own choices. Worth a look.

  • 7

    Very specific

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

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

    If shipright.yaml exists in the project root (or the user passed --check), do NOT run setup.

    line 13
  2. Very specific:

    Ask: "Where should the brain (your project's memory repo) live?

    line 25
  3. Distinctive:

    Ask: "What is this product called?

    line 33
  4. Very specific:

    If you just created the brain: rename products/example-product to products/<product>, replace example-product in INDEX.md and in the …

    line 35
  5. Very specific:

    If you reused a brain and products/<product>/ is missing: copy the template's products/example-product/ in as products/<product>, add a row …

    line 37
  6. Distinctive:

    Ask: "Use these? (yes, or tell me what to add or remove)".

    line 49
  7. Distinctive:

    Write it at the project root, exactly this shape (fill in the answers):

    line 53
  8. Distinctive:

    Run the Step 0 checks against what you just wrote (skip the hooks line: they load on the next session start).

    line 72
  9. Distinctive:

    Ask: "Want a 20-second demo of the gate?

    line 76
  10. Very specific:

    Create a scratch file whose path matches a red zone, for example migrations/0001_shipright_demo.py, containing a one-line comment.

    line 79
  11. Very specific:

    Ask once: "Want to see Level 2 (modes and delegation)?" If yes, describe /shipright:mode and /shipright:delegate in two sentences each and …

    line 86
  12. Very specific:

    Never edit or delete anything outside the brain folder and shipright.yaml, except the demo scratch file.

    line 92
  13. Distinctive:

    Never push anywhere. The brain stays local until the user adds a remote.

    line 93
  14. Distinctive:

    If the user is unsure at any step, take the default and say what you chose.

    line 94

“Common” is measured: kept by at least 6 in 10 of the 171 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

/shipright:setup

You are guiding the user through setup in a conversation. Ask one question at a time, offer a default for each, and wait for the answer. Plain English; gloss any term once ("red zone = a path where edits need a signed decision first").

Step 0: check mode

If shipright.yaml exists in the project root (or the user passed --check), do NOT run setup. Instead check and print a checklist with a tick or cross per line, then stop:

  • Config parses and has brain, product, red_zones, banks, models.
  • The brain path exists and contains GOVERNANCE.md.
  • products/<product>/ exists in the brain and has DIRECTION.md.
  • Lint passes: python3 "${CLAUDE_PLUGIN_ROOT}/scripts/brain_lint.py" <brain> (errors are crosses, warnings are notes).
  • Hooks are loaded: the Shipright banner or gate message appeared this session. If you cannot tell, say so.

For each cross, say in one line how to fix it. If the config is broken, offer to rerun setup from Step 1.

Step 1: where the brain lives

Ask: "Where should the brain (your project's memory repo) live? Default: ../brain, next to this project."

  • Folder does not exist: copy ${CLAUDE_PLUGIN_ROOT}/brain-template/ there, run git init inside it, and continue. Do not commit yet (Step 2 renames things first).
  • Folder exists and has GOVERNANCE.md: it is a brain. Reuse it and say so.
  • Folder exists but is not a brain: stop and ask for a different path.

Step 2: product name

Ask: "What is this product called? Default: the project folder name." Use lowercase letters, digits and dashes.

If you just created the brain: rename products/example-product to products/<product>, replace example-product in INDEX.md and in the DIRECTION.md, MAP.md and log.md headings, and set the first log entry's date to today. Then commit with the message "Brain created by Shipright".

If you reused a brain and products/<product>/ is missing: copy the template's products/example-product/ in as products/<product>, add a row to INDEX.md, and commit "Add <product>".

Step 3: red zones

Say: "Red zones are paths where I will refuse to edit until a signed decision exists for the current branch. Defaults:"

**/migrations/**
**/payment/**
**/billing/**

Ask: "Use these? (yes, or tell me what to add or remove)". One "yes" accepts all three. Turn plain-language answers ("anything touching invoices") into globs and read them back for confirmation.

Step 4: write shipright.yaml

Write it at the project root, exactly this shape (fill in the answers):

brain: <brain path>        # relative to project root or absolute
product: <product>
red_zones:
  - "<glob>"
gate: commit
banks: [core]
models:
  judge: inherit
  executor: sonnet
  scout: haiku

Then say one sentence: "gate: commit means a consensus doc only opens the gate once a human has committed it in the brain; set gate: file to relax that." Do not ask a question about it.

Step 5: self-check

Run the Step 0 checks against what you just wrote (skip the hooks line: they load on the next session start). Print the checklist. If any line fails, fix it before moving on.

Step 6: demo (ask first)

Ask: "Want a 20-second demo of the gate? I will try to edit a red-zone file so you see a block once."

If yes:

  1. Create a scratch file whose path matches a red zone, for example migrations/0001_shipright_demo.py, containing a one-line comment.
  2. Try to edit it with the Edit tool. Expect a block message. If it is not blocked, tell the user the hooks are not active yet and that restarting the session may be needed.
  3. Show the block message and delete the scratch file (and the folder if you created it).
  4. Explain in two sentences: the gate stops edits to risky paths until a signed decision exists for the current branch, and it can only say no, never approve. There are two exits (in the default commit tier the decision doc must also be committed by a human in the brain): run /shipright:grill to produce the signed decision, or override with a recorded reason via /shipright:grill --override '<reason>'.

Step 7: finish

Ask once: "Want to see Level 2 (modes and delegation)?" If yes, describe /shipright:mode and /shipright:delegate in two sentences each and stop. If no, end with exactly:

"You're set. Next: /shipright:grill before any red-zone change; /shipright:where whenever you're lost."

Rules

  • Never edit or delete anything outside the brain folder and shipright.yaml, except the demo scratch file.
  • Never push anywhere. The brain stays local until the user adds a remote.
  • If the user is unsure at any step, take the default and say what you chose.

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.