agentleFS
Sign inStart free
karloowskarloows/orchraft/skills/warpath/SKILL.md

warpath

Skill1 starsChanged 52 days agoApache-2.0View on GitHub

Run the fix, ship, and roast cycle for an open pull request until the latest roast reports zero findings. Use when the user asks to repeat review fixes until clean or run the review loop.

The file

We split this skill into the rules it gives an agent

9 rules. About 0 are common, 3 are distinctive, and 6 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈3

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 6

    Very specific

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

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

    Use this skill when the user wants Orchraft to keep cycling through review findings instead of stopping after one ship or roast pass.

    line 8
  2. Very specific:

    Read the live branch and pull-request state with runes.

    line 13
  3. Very specific:

    If findings exist, hand them to Forgehand for fixes.

    line 14
  4. Very specific:

    When fixes are ready, hand off to Raid Captain / ship.

    line 15
  5. Very specific:

    Run roast against the current pull-request head, including the initial head when no prior roast exists.

    line 16
  6. Distinctive:

    If findings remain, repeat from step 2.

    line 18
  7. Distinctive:

    If a finding is declined, keep it visible and stop for the user's decision.

    line 27
  8. Very specific:

    Read context/policies/approval-policy.md before starting.

    line 33
  9. Distinctive:

    Do not create a stored loop state file.

    line 40

≈ “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

Warpath Workflow

Use this skill when the user wants Orchraft to keep cycling through review findings instead of stopping after one ship or roast pass.

At A Glance

  1. Read the live branch and pull-request state with runes.
  2. If findings exist, hand them to Forgehand for fixes.
  3. When fixes are ready, hand off to Raid Captain / ship.
  4. Run roast against the current pull-request head, including the initial head when no prior roast exists.
  5. If findings remain, repeat from step 2; otherwise the latest roast is clean.
  6. Recommend Haulmaster / land; do not merge as part of this loop.

Completion

The loop is complete only when the latest roast pass ran against the current PR head and reported zero findings. A successful ship, passing checks, or zero unresolved threads from an older head is not completion.

If a finding is declined, keep it visible and stop for the user's decision; declining is not zero findings. If checks fail, the PR disappears, or the review cannot be posted, stop and report the blocking state.

Approval

Read context/policies/approval-policy.md before starting. This workflow is an orchestrator, not a permission grant. If the target repo's verified Autonomous Mode names the action under autonomous.warpath, continue through that action without asking again during this user-started run. Otherwise pause before every commit, push, PR update, review, thread resolution, or other mutation. Landing always remains a separate, explicit user request.

Do not create a stored loop state file. Re-read live git and GitHub state after each handoff so a resumed run cannot mistake a stale roast for a clean one.

Handoff

Unless the user asked for plain output, open with a short orc-voice line, speaking as the Warpath from context/personality.md (the target repo's copy when it exists, otherwise ${CLAUDE_PLUGIN_ROOT}/context/personality.md), and state the current cycle. Report every cycle's finding count and PR head. When clean, hand off to Haulmaster / land instead of invoking it.

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