multi-agent-work-coordinator

A planning and oversight guide for splitting repository work among multiple coding agents. It maps dependencies, assigns exclusive ownership of files or areas, detects overlap, and defines how results are integrated and checked.

In plain words
What is it for?
Use it to divide independent changes into coordinated work items, order dependent tasks, review agent results, and verify acceptance criteria.
Why use it?
It reduces conflicting edits, unsafe parallel work, and unclear responsibility when several agents work in one code repository.

Skill for Claude CodeCodex

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add skills/cmdr-chara/codex-toolkit/multi-agent-work-coordinator
Any agent
npx skills add cmdr-chara/codex-toolkit --skill multi-agent-work-coordinator
Clone the repo
git clone --depth 1 https://github.com/cmdr-chara/codex-toolkit

Made for: Claude Code, Codex.

Per session 78 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,763 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
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 $0.00078 $0.01763
Opus 5 $0.00039 $0.00881
Sonnet 5 $0.00016 $0.00353
Haiku 4.5 $0.00008 $0.00176

Measured 2d ago against content hash cf917f25827b, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

multi-agent-work-coordinator 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 2d ago.

The scan reads SKILL.md. This mod also ships 1 executable file (scripts/ownership_check.py), listed below but not scanned — reading those needs a real analyzer, not pattern matching.

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.

skills/multi-agent-work-coordinator/SKILL.md · 196 lines

How it starts

The opening of the file, as written. The whole thing — 196 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Multi-Agent Work Coordinator

Parallelize execution without parallelizing accountability. The parent owns the work graph, integration, validation, and final judgment.

Trigger boundary

Use this skill for:

  • splitting a mapped feature across independent packages, tests, and documentation;
  • coordinating multiple agents while preventing shared-file collisions;
  • sequencing migrations or generated outputs through explicit dependency waves;
  • auditing a proposed agent plan for overlap, missing acceptance evidence, or unsafe integration.

Do not trigger when:

  • repository boundaries and consumers are unknown—run repository-intelligence first;
  • one agent can complete the work with less briefing and review overhead;
  • tasks require continuous edits to the same files or unresolved product decisions;
  • the request is only to choose specialized Mission Control roles—the existing delegate-with-mission-cards skill can dispatch after this skill defines the work graph.

Required inputs

Require:

  1. one objective and concrete acceptance criteria;
  2. current repository state and an intelligence map or equivalent evidence;
  3. candidate work items with likely read/write surfaces;
  4. single-owner surfaces, sequence constraints, and integration checks;
  5. available agent capabilities and any access restrictions.

If the repository map is stale relative to the working tree, refresh the affected slice before assigning work.

Safety baseline

  • Inspect and preserve uncommitted user work.
  • Do not use Git cleanup, force operations, destructive migrations, or unreviewed generated output.
  • Give every writer an exclusive write scope. “Coordinate later” is not a valid collision strategy.
  • Keep lockfiles, schemas, migrations, central registries, generated artifacts, release metadata, and shared configuration under one writer or the parent integrator.
  • Agents may read outside their write scope only as stated. They must stop before editing outside it.
  • Never accept completion solely from an agent's prose; inspect artifacts and rerun proportionate checks in the integrated state.

Read the full file on GitHub · 196 lines

Files

What ships with it

4 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.

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. 2d ago First seen · 196 lines · 78 tokens per session scan A cf917f25827b

Subscribe to this mod's changes

multi-agent-work-coordinator is a skill published in the GitHub repository cmdr-chara/codex-toolkit (2 stars, last pushed 5d ago), licensed MIT. It adds 78 tokens to every session and 1,763 once invoked, about $0.0004 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 2026-08-31.

Related

Other skills, from other repositories