agentleFS
Sign inStart free
bnemabnema/pi-pstack/skills/setup-pstack/SKILL.md

setup-pstack

Skill2 starsChanged 3 months agoMITView on GitHub

Configure which Pi models pstack uses per role. Detects available provider/model specs and writes pstack's user configuration. Use for /setup-pstack, "configure pstack models", or changing pstack model choices.

We split this skill into the rules it gives an agent

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

  • 0

    Common

    Most of the 24 edited versions we read keep these.

  • 0

    Distinctive

    This author's own choices. Worth a look.

  • 4

    Very specific

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

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

    Write ~/.pi/agent/pstack-models.json.

    line 8
  2. Very specific:

    Run pi --list-models and collect the available provider/model specs.

    line 12
  3. Very specific:

    Read the existing JSON file when present.

    line 16
  4. Very specific:

    If the project has no real-surface verification harness, offer once to run /skill:create-verification-skill.

    line 54

“Common” is measured: kept by at least 6 in 10 of the 24 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

Setup pstack

Write ~/.pi/agent/pstack-models.json. Pstack skills read this file when they delegate through pstack_task. Missing roles inherit the active parent model.

1. Detect models

Run pi --list-models and collect the available provider/model specs. Treat that output as authoritative. inherit-parent and auto are aliases for the active parent model and are always valid.

2. Load current state

Read the existing JSON file when present. Otherwise start with the defaults below. Every scalar role defaults to inherit-parent; every panel defaults to a one-element inherit-parent list. This is portable and avoids inventing unavailable model names.

3. Choose roles

Show the current mapping. Use the available structured question tool when the user wants changes. Offer only detected model specs plus inherit-parent and auto.

Panel arrays set fan-out. One pstack_task child runs per entry. Recommend model-family diversity for review panels when the user has several providers configured, but do not require it.

4. Validate and write

Reject every non-alias value that was not present in pi --list-models. Create the parent directory and overwrite the whole file so reruns are idempotent.

{
  "feature, refactoring": "inherit-parent",
  "bug-fix": "inherit-parent",
  "perf-issue": "inherit-parent",
  "hillclimb": "inherit-parent",
  "judgment and prose": "inherit-parent",
  "hardest tasks": "inherit-parent",
  "how explorer": "inherit-parent",
  "how explainer": "inherit-parent",
  "how critics": ["inherit-parent"],
  "why investigators": "inherit-parent",
  "why synthesizer": "inherit-parent",
  "reflect tooling": "inherit-parent",
  "reflect judgment, divergent, synthesizer": "inherit-parent",
  "arena runners": ["inherit-parent"],
  "arena cross-judge pool": ["inherit-parent"],
  "architect runners": ["inherit-parent"],
  "interrogate reviewers": ["inherit-parent"]
}

Tell the user which path was written. New pstack_task calls use it immediately because the skills read configuration at invocation time.

Optional verification skill

If the project has no real-surface verification harness, offer once to run /skill:create-verification-skill. Do not push after a refusal.

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