swarm

Coordinate parallel task execution across lanes with workspace policies and merged results.

3|1|Updated Mar 23, 2026

Search Tags:#automation #planning #parallel #swarm #coordination #workflows #lanes

Ready to run

npx skills add https://github.com/malinskibeniamin/skills --skill swarm-malinskibeniamin

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: swarm
Source: https://github.com/malinskibeniamin/skills/tree/main/swarm
Command: npx skills add https://github.com/malinskibeniamin/skills --skill swarm-malinskibeniamin

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Orchestrates parallel task execution across lanes to accelerate complex project work.

Core Features & Use Cases

Lane-based parallel execution with policy-driven workspace management and isolation.
Coordinated kickoff, result merging, and conflict resolution to avoid interference.
Context propagation and structured outputs (status, summary, findings, next_action) across lanes.
Use Case: accelerate codebase changes by launching independent lanes for refactors, experiments, and validations, then merge results.

Quick Start

Provide a swarm manifest describing lanes, workspace_policy, and goals, then initiate the swarm to execute parallel lanes.

Frequently Asked Questions about swarm

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

FAQPage Schema

How do I coordinate parallel task execution across multiple workstreams?▼

Parallel task execution is coordinated by defining a swarm manifest with separate lanes, workspace policies, and goals. The system orchestrates isolated execution across these lanes, merging results and resolving conflicts to accelerate complex project work.

What is a swarm manifest and how does it manage parallel workflows?▼

A swarm manifest is a structured input defining parallel execution lanes, workspace isolation policies, task scope, and output schemas. It manages workflows by propagating context across lanes and enforcing guardrails to prevent conflicting codebase changes.

How do I prevent conflicting changes when running parallel refactoring tasks?▼

To prevent conflicting changes during parallel tasks, the system applies policy-driven workspace management and isolation. It uses coordinated kickoff and result merging guardrails defined in the swarm manifest to avoid interference across lanes.

Can I use lane-based execution for isolated codebase experiments and validations?▼

Lane-based execution supports isolated codebase experiments, refactors, and validations by launching independent lanes. It provides structured outputs including status, summaries, findings, and next actions, then merges the finalized results.

What do I need to provide to start parallel lane execution for a complex project?▼

To start parallel lane execution, you must provide a swarm manifest containing lane definitions, a workspace policy, task scope, and output schema. The system uses this context to generate a finalized plan with execution status and findings.