skills.rest
JMTassy
JMTassyCOMMUNITY
Explore all skills by this author →
GitHub

ops/run_coordination

Coordinate multi-agent runs with shared roots and isolated worker lanes.

Updated Apr 4, 2026
Search Tags:#multi-agent#provenance#coordination#work-lanes#run-protocol
One-click install
npx skills add https://github.com/JMTassy/helen-conquest --skill ops-run-coordination
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: ops/run_coordination
Source: https://github.com/JMTassy/helen-conquest/tree/main/oracle_town/skills/ops/run_coordination
Command: npx skills add https://github.com/JMTassy/helen-conquest --skill ops-run-coordination

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Coordinate multi-agent contributions by establishing a shared run root and isolated worker lanes.

Core Features & Use Cases

  • Run root with per-agent lanes to isolate narratives and ensure a single synthesis surface at the root.
  • Canonical directory layout including briefs, tasks, decisions, and per-worker outputs to preserve provenance.
  • Clear promotion and retrofit guidelines to lift patterns into doctrine and maintain auditability.

Quick Start

Create a new run directory under ops/runs with a unique run-id, initialize brief.md and tasks.md, and establish workers/<lane>/output.md directories for each agent-machine pair.

Frequently Asked Questions about ops/run_coordination

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I coordinate multi-agent contributions across isolated worker lanes?▼

Multi-agent coordination is managed by establishing a shared run root with isolated worker lanes, prescribing run-id naming, directory layouts, and append-only artifacts to preserve provenance across machines.

What is the standard directory layout for multi-agent run provenance?▼

A provenance directory layout uses a run root containing brief.md, tasks.md, decisions.md, and final.md, alongside dedicated per-worker output logs to separate narratives and maintain traceability.

How do I initialize a new multi-agent run with isolated lanes?▼

To initialize a multi-agent run, create a directory under ops/runs with a unique run-id, populate brief.md and tasks.md, and set up workers/<lane>/output.md paths for each agent-machine pair.

Can I use append-only artifacts for multi-agent traceability across multiple machines?▼

Yes, append-only artifacts enforce non-sovereign writes in run artifacts, ensuring that multiple machines and agents maintain strict traceability and isolated narratives within their dedicated files.

What's the best way to synthesize multi-agent outputs into a single audit trail?▼

The best way to synthesize multi-agent outputs is using a single synthesis surface at the run root, combining per-agent worker lanes and canonical files like decisions.md and final.md for auditability.

Related Skills

Explore More Software Engineering Skills→
MadAppGangMadAppGang
278

claudemem-orchestration

Coordinate parallel multi-agent code analyses by sharing claudemem outputs across sessions.

Official
Advanced
thongdn-itthongdn-it
3

vercel-composition-patterns

Replace boolean prop proliferation with Vue composition patterns.

Community
Advanced
kunhuang12345kunhuang12345
1

tmux-workflow

Coordinate Codex workers in tmux sessions with isolated CODEX_HOME directories.

Community
Advanced
gabriellpequenogabriellpequeno
8

parallel-agents

Coordinate multi-agent decomposition with shared context into an integrated report.

Community
Advanced