Lightweight Team Orchestration

Facilitates simple multi-agent team collaboration with defined roles, task distribution, and versioned artifact management via a streamlined folder structure.

Security Scan

The skill is an instruction-only orchestration pattern that is internally consistent with its stated purpose and does not request credentials, installs, or unrelated capabilities.

Purpose & Capability

Name/description (lightweight team orchestration) matches the instructions: creating agent role files, distributing tasks, and versioning artifacts. The skill does not ask for unrelated binaries, environment variables, or cloud credentials.

Instruction Scope

Runtime instructions are scoped to creating folder structure, role definition files (SOUL.md), launching agents via the platform primitive 'sessions_spawn', and saving outputs under projects/. This stays within the described purpose. Note: it instructs use of a 'high-reasoning model' and the sessions_spawn primitive — both are platform-level capabilities that enable autonomous agent launches; this is expected for a multi-agent orchestration skill but you should confirm platform-level session/permission controls before running.

Install Mechanism

No install spec, no code files, and no downloads — lowest-risk installation footprint (instruction-only).

Credentials

No environment variables, credentials, or config paths are requested. This is proportionate to the stated functionality.

Persistence & Privilege

always is false and the skill does not request persistent system presence or modify other skills' configs. Autonomous invocation via the agent is possible (platform default) but not escalated by the skill itself.

Assessment

This skill appears coherent and low-risk: it only describes folder layout, role files, and using the platform's sessions_spawn primitive to run agents. Before installing, confirm you trust the platform's agent/session permissions (sessions_spawn can launch autonomous agents) and enforce limits on what spawned agents can access (secrets, external networks, or system files). Review any SOUL.md role content you create to avoid embedding sensitive data, and ensure your platform enforces per-session access controls and logging.

A lightweight multi-agent team collaboration skill with a simplified output structure focused on core deliverables.

Use Cases

• Creating simple teams with 2+ agents
• Agent role definition and task distribution
• Agent output artifact management (with versioning)
• Lightweight collaboration workflows (no complex workflow engine required)

Quick Start

Output Structure

[TEAM_NAME]/
├── agents/                   # Agent role definitions
│   ├── orchestrator/         # Orchestrator
│   │   └── SOUL.md           # Role definition
│   ├── builder/              # Executor
│   │   └── SOUL.md
│   └── reviewer/             # Reviewer (optional)
│       └── SOUL.md
└── projects/                 # Output artifacts (with versioning)
    ├── v1.0.0/               # Version directory
    │   ├── builder.md        # Builder output
    │   └── reviewer.md       # Reviewer output
    └── v1.1.0/               # Next version

Workflow

1. Create Agent Roles → Create SOUL.md for each agent under agents/
2. Distribute Tasks → Launch agents using sessions_spawn
3. Collect Artifacts → Store artifacts in projects/v{version}/
4. Version Management → Create a new version directory for each delivery

Agent Role Definitions

agents/orchestrator/SOUL.md

# Orchestrator

## Responsibilities
- Task distribution and progress tracking
- Coordination between agents
- Artifact version management

## Behaviors
- Use high-reasoning model
- Keep task status updated

agents/builder/SOUL.md

# Executor

## Responsibilities
- Execute specific tasks
- Produce deliverables

## Behaviors
- Deliver according to specifications
- Annotate version number

agents/reviewer/SOUL.md (optional)

# Reviewer

## Responsibilities
- Verify output quality
- Propose improvement suggestions

Task Distribution

Launch agents using sessions_spawn:

Task: [Task description]
Output: projects/v{version}/{agent}.md
Verify: [Verification method]

Version Management

• Initial version: v1.0.0
• Each iteration: Increment version number
• Preserve history: All versions retained in projects/

When to Use

• Simple 2-3 person teams
• No complex workflow engine required
• Focus on artifact delivery over process

When Not to Use

• Complex multi-stage workflows
• Real-time status tracking → Use task boards
• Large-scale agent coordination