loops / 5dive-ai / loops / multi-agent-task-coordinator

Multi-Agent Task Coordinator

opsimported
Installation
$npx agenticloops install multi-agent-task-coordinator
Summary

Receives tasks from Slack, GitHub, or email; delegates to isolated worker agents; compounds shared memory over time.

  • Runs On inbound task, on any harness.
  • Uses skills: task-delegation, planning, summarize, monitoring.
  • Needs cli docker, secrets SLACK_BOT_TOKEN, GITHUB_TOKEN — checked at install; secrets are prompted, never in the file.
  • Runs at the standard capability tier (balanced model); the harness picks the model.
What it does
Monitor inbound task channels (Slack, GitHub issues, email). When a task arrives, analyze scope, spawn isolated Docker worker agents for each sub-task, track progress via shared persistent memory that compounds context over time, and post a summary to the originating channel when complete.
LOOP.md
---
name: multi-agent-task-coordinator
description: Multi-Agent Task Coordinator · Receives tasks from Slack, GitHub, or email; delegates to isolated worker agents; compounds shared memory over time.
event: slack-message | github-issue | email            # On inbound task
skills: [task-delegation, planning, summarize, monitoring]
requires:
  cli: [docker]
  secrets: [SLACK_BOT_TOKEN, GITHUB_TOKEN]      # names only
tier: standard            # frontier | standard | fast
effort: medium
concurrency: skip
tags: [ops, multi-agent, coordination, automation]
license: MIT
---

Monitor inbound task channels (Slack, GitHub issues, email). When a task arrives, analyze scope, spawn isolated Docker worker agents for each sub-task, track progress via shared persistent memory that compounds context over time, and post a summary to the originating channel when complete.
Installs
tracking soon
Adapted from
desplega-ai/agent-swarm ↗
original author: desplega-ai
GitHub Stars
☆ 558
First Seen
Jun 30, 2026
License
MIT
Trigger
⚡ On inbound task