multi-agent-coordination

Multi-Agent Coordination Skill

Safety Notice

This listing is imported from skills.sh public index metadata. Review upstream SKILL.md and repository scripts before running.

Copy this and send it to your AI assistant to learn

Install skill "multi-agent-coordination" with this command: npx skills add culpritchaos/interlock/culpritchaos-interlock-multi-agent-coordination

Multi-Agent Coordination Skill

Use this skill to keep work synchronized between Antigravity and Claude Code.

When to Use

  • Starting a new session (check current handoff state)

  • Completing a task (update status and next steps)

  • Switching agents (write handoff note)

  • Logging significant decisions or actions

Coordination File

  • Template (public): multi_agent_coord.template.md

  • Working file (local): docs/MULTI_AGENT_COORD.md

The working file is gitignored. It contains sensitive task context that should not be public.

Session Start Protocol

  • Read the working coordination file

  • Check git status for any conflicts

  • Note the current handoff state

  • If another agent was active, acknowledge the handoff

Handoff Protocol

When ending a session or switching agents, update the working file:

Current Handoff State

Last Updated: [YYYY-MM-DD HH:MM TZ] Active Agent: [Your agent name] Status: [What you just completed] Next Steps: [What the next agent should do] Blockers: [Any issues or "None"]

Task Queue Management

Add tasks with priority:

PriorityTaskAssigned ToStatus
1[High priority task]Claude Codein_progress
2[Medium priority task]Antigravitypending

Session Logging

Log significant actions:

DateAgentActionOutcome
2026-01-10Claude CodeRefactored auth moduleComplete

Conflict Resolution

If git status shows changes from both agents:

  • Do NOT overwrite the other agent's work

  • Run git stash to save your changes

  • Report the conflict to Rob

  • Wait for resolution before continuing

One Owner Rule

Only ONE agent should edit the coordination file per session. If you see recent edits from the other agent, read but don't write until handoff is complete.

Source Transparency

This detail page is rendered from real SKILL.md content. Trust labels are metadata-based hints, not a safety guarantee.

Related Skills

Related by shared tags or category signals.

Automation

interlock-ops

No summary provided by upstream source.

Repository SourceNeeds Review
General

maintainer

No summary provided by upstream source.

Repository SourceNeeds Review
Automation

multi-agent-coordination

No summary provided by upstream source.

Repository SourceNeeds Review
Automation

clinic-visit-prep

帮助患者整理就诊前问题、既往记录、检查清单与时间线,不提供诊断。;use for healthcare, intake, prep workflows;do not use for 给诊断结论, 替代医生意见.

Archived SourceRecently Updated