Agent spawner

Skill Razaib-khan/ForgeWeave/src/forgeweave/templates/qwen/.qwen/skills/agent-spawner

Behavioral execution framework for AI agents — define deterministic, portable skills & agents across OpenCode, Claude Code, Gemini CLI, and Qwen Code.

Install

npx -y skills add Razaib-khan/ForgeWeave --skill agent-spawner

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

One thing to look at

1 stars1 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

What its author says it does

Copied from the file, not written here

Creates and configures subagents dynamically with defined roles, constraints, and tool permissions

SKILL.md

2.5 KB, as published. Nobody here has run it

Agent Spawner

Purpose

Dynamically create and configure sub-agents for specific tasks. Each spawned agent gets a defined role, constraints, tool permissions, and a task-specific prompt. Enables the multi-agent patterns used in deep-research and workflow-orchestrator.

When to Use

A multi-agent pipeline needs to be executed (e.g., deep-research)
A task requires a specialized agent with specific permissions
Parallel execution of independent subtasks is needed
An agent needs to be sandboxed with restricted capabilities

When Not to Use

The task is simple enough for a single agent
The user explicitly asked for a single-agent approach
The overhead of spawning a sub-agent exceeds the complexity of the task

Inputs

Input	Type	Required	Description
`task`	string	Yes	What the sub-agent should do
`role`	string	Yes	The sub-agent's role (e.g., "researcher", "validator")
`constraints`	array	No	Rules the sub-agent must follow
`permissions`	object	No	Tool permissions for the sub-agent

Expected Outputs

Output	Description
Spawned agent	A configured sub-agent ready to execute
Agent manifest	The agent's instructions and constraints

Exact Workflow Steps

Define the agent's role based on the task requirements
Set constraints: what the agent MUST and MUST NOT do
Set permissions: which tools the agent can use
Write the agent's instructions prompt
Spawn the agent using the TUI's sub-agent mechanism
Monitor execution for constraint violations

Required Checks

Agent has a clearly defined role
Constraints are explicit and enforceable
Permissions match the task requirements
Agent instructions are self-contained

Failure Modes

Failure Condition	Response
Permission mismatch	Adjust permissions to match the task
Agent goes out of scope	Enforce constraints and restart
Sub-agent mechanism unavailable	Execute the task directly with a prompt template

References

Reference	Path
Agent Spec	`./AGENT_SPEC.md`
AGENTS.md	`./AGENTS.md`
Deep Research skill	`../deep-research/SKILL.md`