agentleFS
SubliminalCodingSubliminalCoding/bilbroswagginz-skills/skills/qwen-kickstart/SKILL.md

qwen-kickstart

Skill0 starsChanged 47 days agoMITView on GitHub

Local-Qwen 0→1 scaffolder (token-free). Mechanical checklist form of /kickstart for a small local model: write the Intent, pick the SIMPLEST stack, build a walking skeleton that runs end-to-end, list only the features the Intent needs as beads, hand off to qwen-project-autopilot. Anti-overbuild by construction. Use when you want /kickstart's job done on a local model with no API tokens: starting a new project.

The file

We split this skill into the rules it gives an agent

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

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈1

    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:

    No abstraction/framework/config layer until 3 real places actually need it.

    line 50
  2. Distinctive:

    Do NOT run idea-wizard locally to brainstorm many features.

    line 51

≈ “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

qwen-kickstart (local 0→1)

The Qwen-run version of /kickstart. Build the lean scaffold and hand off. Do NOT try to finish the whole project here.

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. Judgment calls belong to the user.
  • ONE change at a time; build → tests → lint after each; green → commit; red → revert, one retry, else STOP. Never weaken tests.
  • Build ONLY what the Intent needs; extras → SOMEDAY.md.
  • Unclear / judgment / outward → DECISIONS.md + STOP.
  • 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.

Steps (in order)

1. INTENT. Write one paragraph: who it's for, the one job, what "done" looks like, keep
   it small. Show the user, wait for "go". This is the CEILING: never build past it.

2. STACK. Pick the SIMPLEST, most boring, most proven way to build it. If the choice is
   big or non-obvious (core framework, data model) → STOP, ask in DECISIONS.md. Default
   to the obvious lean option; do not over-engineer.

3. WALKING SKELETON. Build the THINNEST thing that runs end-to-end (one real path through
   the whole app) + ONE passing test. Commit. This is the baseline the ratchet needs.

4. BEADS. List the features the Intent needs as beads (`br`). One bead per feature. Add
   NOTHING beyond the Intent (extras → SOMEDAY.md).

5. HAND OFF. Report: "scaffold green, beads ready" and hand to qwen-project-autopilot.

Anti-overbuild (these are the point; do not skip)

  • Lean default in step 2; walking skeleton before any deep build in step 3.
  • Intent is the ceiling; beyond-Intent ideas go to SOMEDAY.md, never built.
  • No abstraction/framework/config layer until 3 real places actually need it.
  • Do NOT run idea-wizard locally to brainstorm many features; that's the user's call (overbuild risk on a small model). Build the obvious Intent features only.

Distilled from /kickstart (its references hold the full idea-wizard / advisor-council version for when a human or Claude is driving the 0→1 design).

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.