run

Name: run
Availability: InStock
Author: stefans71

Coordinate parallel frontier-model agents and synthesize their reports.

5|1|Updated Feb 28, 2026

Search Tags:#automation #cli #agents #review #synthesis #board #frontier

Ready to run

npx skills add https://github.com/stefans71/FrontierBoard --skill run-stefans71

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: run
Source: https://github.com/stefans71/FrontierBoard/tree/main/.claude/skills/run
Command: npx skills add https://github.com/stefans71/FrontierBoard --skill run-stefans71

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

FrontierBoard Run coordinates multiple frontier-model agents to run in parallel and produce a unified synthesis from their reports.

Core Features & Use Cases

Parallel execution of all agents following the board's parallelism rules.
Inbox readiness checks with prompts to run /brief when needed.
Consolidated synthesis of consensus, divergences, conflicts, and overall health.
Logging of findings to board/REVIEW-LOG.md and optional PR submission.

Quick Start

Trigger /run to launch all agents in parallel and generate the board synthesis from their reports.

Frequently Asked Questions about run

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

FAQPage Schema

How do I run parallel agents and synthesize their findings into a single report?▼

To run parallel agents and synthesize findings, trigger the /run command to launch all frontier-model agents simultaneously. The system consolidates their independent reviews into a unified synthesis covering consensus, divergences, conflicts, and overall health.

What is parallel agent synthesis for project reviews?▼

Parallel agent synthesis coordinates multiple frontier-model agents running independently to answer a project question. It merges their separate reports into a unified summary that highlights consensus, divergences, and conflicts, providing comprehensive health insights.

Do I need a board.md file to coordinate parallel frontier-model agents?▼

Yes, coordinating parallel frontier-model agents requires access to board.md and per-agent in/out boxes. The board file defines the parallelism rules that the agents follow during execution, ensuring proper coordination.

What happens if an agent inbox is not ready before launching parallel execution?▼

If an agent inbox is not ready before launching parallel execution, the system provides prompts to run the /brief command. This ensures all agents have the necessary context and inputs before they begin their independent reviews.

How are synthesized findings from parallel agents logged and submitted?▼

Synthesized findings from parallel agents are logged to board/REVIEW-LOG.md. The system also offers optional PR submission to integrate the consolidated synthesis of consensus and divergences directly into the project workflow.