agentleFS
Sign inStart free
SubliminalCodingSubliminalCoding/bilbroswagginz-skills/skills/qwen-ship-it/SKILL.md

qwen-ship-it

Skill0 starsChanged 47 days agoMITView on GitHub

Local-Qwen release pre-flight (token-free). Mechanical form of /ship-it: do ONLY the reversible pre-flight (confirm green + intent, bump version, changelog, build artifacts + checksums, verify installer), then list every outward button (push/publish/deploy/ submit) with its rollback in DECISIONS.md and PRESS NOTHING. Use when you want /ship-it's job done on a local model with no API tokens: preparing a release while the buttons stay with the user.

The file

Create a free account to mount
Files
SKILL.md

qwen-ship-it (local release pre-flight)

The Qwen-run version of /ship-it. The whole skill is "do the safe pre-flight, then STOP at every button." A small local model must NEVER press an outward action.

QWEN LOCAL RULES (full text: ../qwen-forge/references/QWEN-MODE.md)

Requires qwen-forge installed alongside this skill (it holds the shared rules).

  • Builder, not architect. ONE change at a time; build → tests → lint; green → commit; red → revert, one retry, else STOP. Never weaken tests.
  • Local endpoint required: a local OpenAI-compatible endpoint (vLLM, SGLang, llama.cpp or Ollama) serving a capable Qwen-class coding model. ONE model loaded, work SEQUENTIALLY, leave memory headroom on unified-memory machines.
  • ABSOLUTE: never push, publish, deploy, submit, or run cargo/npm publish. Those are buttons.

Steps (reversible pre-flight only)

1. GATE. Confirm build + tests are green AND the project delivers its Intent. If not →
   STOP, report what's missing. Do not prepare a release of something broken/incomplete.

2. SCAN. Check the artifacts/config for secrets/keys. Found one → STOP, DECISIONS.md.

3. PREP (all reversible, commit as you go):
     - bump the version (semver).
     - write the changelog / release notes.
     - build the release artifacts + checksums locally.
     - verify the installer actually installs/runs, if there is one.
     - deploy to a PREVIEW/staging env only, never production.

4. BUTTONS. Write to .autopilot/DECISIONS.md every outward action, each with its exact
   command and its rollback:
     - push tag / release · publish GH release · promote preview→production ·
       submit to store · cargo/npm publish
   PRESS NONE OF THEM. Report the ready package + the button list to the user.

Rule

  • Everything reversible and local = you do it. Everything outward/irreversible = a button in DECISIONS.md for the user. When unsure which, it's a button.

Distilled from /ship-it. Its ROUTING.md holds the full button-discipline + platform- decoupling version for a Claude/human driver.

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