ac-workflow-mux-subagent

ac-workflow-mux-subagent is a skill for Claude Code from WaterplanAI/agentic-config. It costs 34 tokens per session (697 once invoked), scanned A, original, MIT.

A worker protocol for MUX subagents, which are helper agents coordinated by a parent agent. It defines how workers report results, signal completion, and return status.

In plain words
What is it for?
Use it when implementing work as a subagent and when writing required reports and completion signals.
Why use it?
It gives the coordinator consistent evidence that a worker finished successfully or is blocked.

Skill for Claude Code

Written for Claude Code: allowed-tools in frontmatter. Also seen: mentions subagents.

Needs its repository: it reads a path above its own folder, which exists only inside the repository. The line is uv run ../../assets/mux/tools/signal.py <signal-path> --path <report-path> --status success.

Part of the agentic-config plugin — 49 skills, 1 plugin shipped together

Good fit Use it when implementing work as a subagent and when writing required reports and completion signals.

Compare 6 skills from other repositories ↓
Install

Getting it into your agent

It runs from inside its repository, so the clone comes first — what it calls does not travel with the file alone.

Clone the repo
git clone --depth 1 https://github.com/WaterplanAI/agentic-config
agentmods
npx agentmods add skills/waterplanai/agentic-config/ac-workflow-mux-subagent

Made for: Claude Code.

Or install agentic-config, the plugin that ships this one along with the rest of its 49 skills, 1 plugin.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for ac-workflow-mux-subagent

README.md
[![agentmods](https://agentmods.dev/badge/skills/waterplanai/agentic-config/ac-workflow-mux-subagent/github.svg)](https://agentmods.dev/skills/waterplanai/agentic-config/ac-workflow-mux-subagent)
Your own site
<a href="https://agentmods.dev/skills/waterplanai/agentic-config/ac-workflow-mux-subagent"><img src="https://agentmods.dev/badge/skills/waterplanai/agentic-config/ac-workflow-mux-subagent/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

agentmods 80×15 button for ac-workflow-mux-subagent

Your own site · 80×15
<a href="https://agentmods.dev/skills/waterplanai/agentic-config/ac-workflow-mux-subagent"><img src="https://agentmods.dev/badge/skills/waterplanai/agentic-config/ac-workflow-mux-subagent.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 34 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 697 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5.1 $0.00034 $0.00697
Opus 5 $0.00017 $0.00349
Sonnet 5 $0.00007 $0.00139
Haiku 4.5 $0.00003 $0.00070

Measured 10d ago against content hash bf78bf8440e8, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-09, from the pricing page.

Security

Grade A, and why

ac-workflow-mux-subagent scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 10d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

packages/pi-ac-workflow/skills/ac-workflow-mux-subagent/SKILL.md · 90 lines

How it starts

The opening of the file, as written. The whole thing — 90 lines — stays where its author put it; the contents beside it link to each section on GitHub.

MUX Subagent Protocol

Purpose

This is the pi-adapted worker contract for mux.

Binding activation

If the parent coordinator explicitly invokes ac-workflow-mux-subagent, or embeds this skill text as the worker runtime for the task, treat this document as a binding runtime contract, not as optional guidance, commentary, or a planning reference.

Workers stay data-plane only and do not own control-plane progression. Coordinators own ledger transitions and strict gate progression.

Companion protocol artifacts:

  • ../../assets/mux/protocol/subagent.md
  • ../../assets/mux/protocol/guardrail-policy.md
  • ../../assets/mux/protocol/strict-happy-path-transcript.md
  • ../../assets/mux/protocol/strict-blocker-path-transcript.md
  • ../../assets/mux/protocol/strict-regression-checklist.md

Runtime Differences From the Source Runtime

  • There is no TaskOutput tool in current pi runtime.
  • There is no nested Skill(...) loader inside a pi worker.
  • Worker depth stays at one layer: coordinator -> subagent.
  • Completion is verified through report/signal artifacts plus summary evidence gating.

Mandatory Worker Rules

  • Write all substantive results to the report file path provided by the parent coordinator.
  • Create a success or failure signal before finishing.
  • Keep the final textual response exactly 0 on success.
  • Do not launch nested subagent calls.
  • Do not call control-plane bridge tools or report_parent; mux workers communicate through report/signal artifacts only.
  • Do not mutate control-plane state (ledger.py transition, coordinator dispatch mutation, or manual ADVANCE fallback).
  • Put routing guidance in the report Executive Summary so the coordinator can route the next wave efficiently.

Signal Command

uv run ../../assets/mux/tools/signal.py <signal-path> --path <report-path> --status success

Report Format

# <Report Title>

## Table of Contents
(all markdown headers)

## Executive Summary
- **Finding/Result**: keyword-dense value
- **Finding/Result**: keyword-dense value
- (max 5-7 bullets)

### Next Steps
- **Recommended action**: what the coordinator should do next
- **Dependencies**: files or reports this connects to
- **Routing hint**: which worker type should consume this next

## <Sections>
(your detailed content)

Read the full file on GitHub · 90 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 10d ago First seen · 90 lines · 34 tokens per session scan A bf78bf8440e8

Subscribe to this mod's changes

ac-workflow-mux-subagent is a skill published in the GitHub repository WaterplanAI/agentic-config (30 stars, last pushed 1mo ago), licensed MIT. It adds 34 tokens to every session and 697 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.