piflow

Pi Flow — author, enhance, and run structured filesystem-coordinated workflows (a DAG of producer/verify nodes) as a fleet of efficient pi agents driven by non-Claude coding-plan models, with Claude Code as the single console. This is the plugin OVERVIEW/router; the work lives in three skills — piflow-init (create), piflow-enhance (improve), piflow-start (run). Source of truth is a structured workflow TEMPLATE; the `@piflow/core` SDK loads it and runs one `pi` per node.

Jump to install

Repository: blueif16/PiFlow
Last source activity: July 24, 2026 at 21:05
Detected SKILL.md language: English
Stars: 97
Forks: 1

Install options

The review-first prompt is selected by default. You can switch to a direct command or download a local copy.

Install with Codex or Claude Copy this prompt, paste it into Codex, Claude, or another assistant, and let it review the skill page and install it for you.

A direct command skips the review prompt. Inspect the source before running it.

npx skills add https://github.com/blueif16/PiFlow --skill piflow

The command stays on one line. Scroll horizontally to inspect it before copying.

Prefer a local copy? Download the files currently available to SkillsMP.

Review the source files

Read SKILL.md and any companion files shown by SkillsMP before deciding whether to install.

SKILL.md

Source instructions · Read-only preview

Back to search Back to installation

piflow-maintenance

Pi Flow · MAINTENANCE — the canonical STANDARDS reference for a node config + a run condition: the exact per-node lifecycle, the `op[]` action envelope, PRE/POST hooks, gates & policy, how a node FINISHES (fenced-JSON tail · `submit_result` · `returnMode` · verified-not-trusted), run PROFILES, and the token / contract vocabulary. The brother of piflow-init — init decides WHAT the DAG is (divide the purpose, assign the nodes); MAINTENANCE answers "what is our canonical way to DO this?" for an already-built or half-built loop. LOAD/FOLLOW when editing or extending an existing node.json / template and you need the standard: "where does a script go (which op, pre or post)", "how should this node end / return its result", "how do I add a gate / check / judge", "the exact node lifecycle", "how do promote / state channels work", "set up a profile", "inject vs readScope", "is this the right op syntax". REFERENCE, not a loop — carries the standards only, never the optimize loop and never the divide-and-assign creation

2026-07-17

piflow-start

Pi Flow · START — run & monitor an already-created pi-flow workflow on the pi fleet via the SDK CLI, with Claude Code as the single console. LOAD THIS SKILL BEFORE running ANY piflow/pi command — it pins the canonical invocation: the npm-linked global `piflowctl run <templateDir>` bin (NOT `node …/dist/cli.js` nor `pi-runner/run.mjs`, both non-canonical) with `--provider <gw> --thinking low --sandbox local` + `--from`/`--until`. Do NOT reconstruct the command from memory. Triggers — load on ANY of: "run / kick off / start / resume my piflow workflow", "do a live run on pi", "run game-omni on pi", "companion-mode run", "monitor / follow / poll a run", "diagnose a stalled / never-wrote / blocked node", or the words "piflow" / "pi-runner" / "pi fleet" / "pi run" appearing at all. The run is ALWAYS: pull the next prompt from the bank → dry-run (free) → live background → poll. To CREATE or PORT the template first use piflow-init.

2026-07-16

piflow-inspect

Pi Flow · INSPECT — the instrument router for looking INTO a run or a node: performance, cost, timing, context composition, loop/stall anomalies, recurrence. LOAD THIS SKILL BEFORE debugging or analyzing ANY pi-flow run/node — it routes each question to the piflowctl verb that already answers it (status · telemetry · trace · logs · inspect · optimize · memory · understand), so you never hand-parse `.pi/` internals or write a throwaway analysis script for data an instrument already surfaces. Triggers — load on ANY of: "why is this node slow / expensive / looping / stalled", "where did the time/tokens go", "what did the model actually read / see / miss", "how many calls did it make", "dissect / profile / audit this run", "compare two runs of a node", "did this failure happen before", or any intent to open a `.pi/sessions/*.jsonl`, `run.json`, or node log by hand. To RUN a workflow use piflow-start; to judge/act mid-run use piflow-overlord.

2026-07-16

View all from this repository

name

piflow

description

Pi Flow — the plugin overview (router)

One-line model: a structured workflow template (.piflow/<wf>/template/) is the single source of truth; the @piflow/core SDK loads it into a WorkflowSpec and runs it one efficient pi per node, while Claude Code owns the graph and monitors. A Claude .js is an optional one-time INGEST seed, discarded after.

This repo IS the @piflow/core product repo AND the piflow plugin. The agent-facing work is split into three skills (each invokable on its own):

Skill	Role	Status
piflow-init	CREATE a workflow: triage the source (PORT a `.js` / IMPORT n8n / COMPOSE) → build the `template/` → stand up the per-repo runner. The bulk — start here.	full
piflow-enhance	IMPROVE a running workflow: the `hermes-skill-system` capture→route→edit→verify→approve loop, the criteria fixture, Companion Mode.	stub (scope declared)
piflow-start	RUN & monitor a workflow on the pi fleet: dry-run → live → .

piflow

Source facts

Install options

Review the source files

More from this repository

Pi Flow — the plugin overview (router)

More from this repository

Pi Flow — the plugin overview (router)

Product repo map (paths relative to this repo root, ~/Desktop/piflow)

Product repo map (paths relative to this repo root, `~/Desktop/piflow`)