Pg workflow

Skill Exekiel179/pingouin-psych-stats/skills/pg-workflow

Run and resume the stateful Pingouin psychology statistics workflow. Use when the user wants an end-to-end, auditable analysis from intake through screening, analysis, approval, reporting, and archive, or asks to continue an existing analysis run.From its SKILL.md

Install

npx -y skills add Exekiel179/pingouin-psych-stats --skill pg-workflow

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

One thing to look at

0 stars0 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

SKILL.md

2.0 KB, 404 tokens by cl100k_base, as published. Nobody here has run it

PG Workflow

Use this skill as the cross-platform orchestration entry point. Claude Code users may use the /pg-* commands; Codex and other agents should invoke this skill directly.

Load

Read these references in order:

../../references/workflow-contract.md
../../references/archive-contract.md
../../references/intake-checklist.md when starting a new run
../../references/supervision-gates.md before approval
../../references/pingouin-optimization.md before generating Pingouin code

Use ../../scripts/workflow_engine.py for every state transition. Never edit run-manifest.json status by hand.

Dispatch

Find the active run under archive/analysis-runs/, or initialize one with scripts/init_analysis_run.py.
Read the manifest status.
Execute only the next allowed stage from workflow-contract.md.
Save the stage artifacts before advancing the state.
If approval returns REVISE, repair and return to analyzed; if BLOCKED, stop and explain the missing decision or data.
Finish only after workflow_engine.py check <run> passes at complete.

Required Handoff Block

Before each stage, return:

Run: <run-id>
State: <current>
Next: <target state and stage>
Inputs: <files/data available>
Outputs: <files this stage will create>
Guardrail: <highest-risk issue>

Hard Rules

Do not run inferential analysis before screening artifacts exist.
Do not report conclusions before an approved or approved_with_notes state.
Do not claim results from code that has not produced stored numerical output.
Do not copy raw participant data into an archive unless the user explicitly confirms the privacy handling.
Keep the S0-S5 audit line in the final handoff.

What ships with it: 1 file

271 B alongside SKILL.md

agents/

openai.yaml271 B

Gives 0 of the 12 instructions most automation workflows skills give in 404 tokens

Counted across 745 of the 1,008 authors here whose files we hold, read 2026-08-07

Write conventional commit messagesin 36 of 745, across 35 files
Delete branches after mergein 30 of 745, across 21 files
Make atomic commitsin 25 of 745, across 15 files
Write minimal code to pass testsin 22 of 745, across 10 files
Re-snapshot after navigation or DOM changesin 21 of 745, across 13 files
Use try-catch for error handlingin 20 of 745, across 8 files
Run tests before committingin 20 of 745, across 12 files
Write tests before implementationin 20 of 745, across 8 files
Configure branch protection rulesin 19 of 745, across 5 files
Explain the why in commit messagesin 19 of 745, across 9 files
Refactor code while tests remain greenin 19 of 745, across 6 files
Interact with elements using refsin 19 of 745, across 11 files

Said here and by no other author read

Use the workflow engine script for every state transition
Find or initialize the active analysis run
Execute only the next allowed stage
Save stage artifacts before advancing state
Return the handoff block before each stage
Keep the audit line in the final handoff

Grouped from the skills themselves: near-identical wordings counted once, and counted by distinct author, so one author publishing three of these counts once. Length counted with cl100k_base; the agent that loads this file may tokenize it differently.