Setup Pstack logo

Setup Pstack

Community
ScriptedAlchemy
setup-pstack

Configure the Codex models and reasoning efforts pstack uses for delegation roles. Use for setup-pstack, pstack model configuration, or changing pstack's budget.

Overview

PublisherScriptedAlchemy
Repositorypstack-codex
Skill namesetup-pstack
Stars
77
Forks
7
Bundled files
Instructions only
LicenseMIT
Links
  • Markdown instructions

    A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.

  • Works with any LLM

    AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.

  • Self-contained

    Everything the model needs lives in the instructions — no extra files to sync.

  • Open source

    Published by ScriptedAlchemy on GitHub. Read the source before you install it.

Installation

Install the Setup Pstack AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

1

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

  1. Open the app and go to Plugins → Skills.
  2. Choose "Install from GitHub".
  3. Paste the skill folder URL below and confirm.
    https://github.com/ScriptedAlchemy/pstack-codex/tree/main/plugins/pstack/skills/setup-pstack
  4. Enable the skill in any chat where you want it available.
Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.
2

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills
git clone --depth 1 https://github.com/ScriptedAlchemy/pstack-codex.git /tmp/pstack-codex
mkdir -p .claude/skills
cp -r /tmp/pstack-codex/plugins/pstack/skills/setup-pstack .claude/skills/setup-pstack
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Setup Pstack in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads Setup Pstack on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so Setup Pstack is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

  • Loaded only when it is needed

    The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.

  • Switch models mid-chat

    Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

Before following this workflow, read the Codex runtime contract. It defines plugin paths, model configuration, delegation, and persistence.

Setup pstack for Codex

Write ~/.codex/pstack-models.md, the override file every delegating pstack skill reads before it spawns subagents. Missing roles keep the defaults below. This file is configuration for pstack, not a Codex model-provider configuration file.

1. Load current state

Read the existing override file when present, preserving its budget and configured role values. Otherwise start from the defaults in step 4. Drop retired roles absent from step 4 (for example how critics) and list each dropped line when showing the proposed configuration. Determine which Codex models the current host exposes from the current tool metadata or official local model picker. Never invent a model slug. inherit-parent and auto are always valid and both mean to omit the model override.

For roles executed through spawn_agent, its model and effort choices are authoritative. A picker entry or a model available through another harness does not establish that it can be spawned here. Follow CODEX.md when an explicit model requires a fresh context fork.

2. Choose a budget

Use the available user-input tool, or ask one concise question. Preserve the current budget on a rerun unless the user changes it. With no configuration, large matches the skill defaults. Offer:

  • unlimited — max reasoning: each model's highest host-supported effort, up to max.
  • large — xhigh reasoning: xhigh.
  • medium — high reasoning: high.
  • small — medium reasoning: medium.

Apply the selected effort to every real model, including panel entries. Preserve existing model choices, list order, and inheritance aliases; aliases keep both overrides unset. Model and effort are separate fields in Codex. Use the highest supported effort at or below the requested target. If none is available, mark the role as needing a choice. Never change the model to satisfy a budget.

3. Confirm role assignments

Show every role, model, and reasoning effort. Allow inherit-parent or auto for any role. Panel roles are ordered lists; one subagent runs per list entry. List length controls panel size. Keep a user's existing override unless it is unavailable.

If the user asks to repair unavailable assignments, choose replacements from the execution tool's available models using their advertised strengths and the role's purpose. Preserve valid assignments, the budget, and panel order and size; report each replacement. That request authorizes updating the file without another confirmation. Otherwise ask about unavailable choices before writing.

4. Write the override

Overwrite ~/.codex/pstack-models.md atomically so reruns are idempotent. Use this shape:

text
# pstack model configuration for Codex. Delete a role line to use its skill default.
# `inherit-parent` or `auto` omits explicit model and reasoning overrides.
# budget: large (xhigh)
feature, refactoring: gpt-6-luna @ xhigh
bug-fix: gpt-6-luna @ xhigh
perf-issue: gpt-6-luna @ xhigh
hillclimb: gpt-6-luna @ xhigh
judgment and prose: gpt-6-astra @ xhigh
hardest tasks: gpt-6-astra @ xhigh
how explorer: gpt-6.1-sol @ xhigh
how explainer: gpt-6-astra @ xhigh
why investigators: gpt-6.1-sol @ xhigh
why synthesizer: gpt-6-astra @ xhigh
reflect tooling: gpt-6.1-sol @ xhigh
reflect judgment, divergent, synthesizer: gpt-6-astra @ xhigh
arena runners: gpt-6-astra @ xhigh, gpt-6-luna @ xhigh
arena cross-judge pool: gpt-6-astra @ xhigh, gpt-6-luna @ xhigh
swarm workers: gpt-6-luna @ xhigh
architect runners: gpt-6-astra @ xhigh, gpt-6-luna @ xhigh
interrogate reviewers: gpt-6-astra @ xhigh, gpt-6-luna @ xhigh

Apply the budget to the separate effort values before writing. Validate every real model and its effort against the execution tool's available choices. Resolve unavailable assignments as described in step 3; do not write a slug the tool cannot accept.

5. Install or refresh custom agents

Locate this installed plugin's root. Copy codex-agents/poteto-agent.toml and codex-agents/comment-sicko.toml into ~/.codex/agents/, replacing only those two pstack-owned files. Do not alter unrelated agents. Tell the user that new Codex sessions pick them up.

6. Confirm

Report the chosen budget, overrides, config path, and installed agent files. Re-running this skill updates the same files.

Optionally offer $create-verification-skill when the repository has no real-surface test harness. Also offer $create-formal-verification-skill once when the project has no formal verification skill: it checks a key property with a model checker, proof, or code-level verifier under explicit bounds. If the user declines either offer, move on.

If the user wants poteto-mode to persist across tasks in this repository, use the optional scoped AGENTS.md method. Do not alter project instructions unless that persistence was requested.

Frequently asked questions

What does the Setup Pstack AI skill do?

Configure the Codex models and reasoning efforts pstack uses for delegation roles. Use for setup-pstack, pstack model configuration, or changing pstack's budget.

Why use Setup Pstack on TypingMind?

Because you install it once and use it with any model. Setup Pstack is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install Setup Pstack in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/ScriptedAlchemy/pstack-codex/tree/main/plugins/pstack/skills/setup-pstack. TypingMind reads its SKILL.md and installs it as a skill you can enable per chat.

Which AI models can use Setup Pstack?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with Setup Pstack?

As many as you like. As long as a model supports skills, you can use Setup Pstack with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the Setup Pstack AI skill free?

Yes. It is published on GitHub by ScriptedAlchemy under the MIT license. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

View all

Set up your own AI workspace now

Get notified about new features and future giveaways by subscribing to our newsletter 👇