Agent Spawner
Creates and configures subagents dynamically with defined roles, constraints, and tool permissions
Install
$ agentstack add skill-razaib-khan-forgeweave-agent-spawner ✓ scanned · ✓ verified, works with Claude Code, Cursor, and more.
Security review
✓ PassedNo issues found. Passed automated security review. · v0.1.0 How review works →
- ✓ Prompt-injection patterns
- ✓ Secret / credential exfiltration
- ✓ Dangerous shell & filesystem operations
- ✓ Untrusted network calls
- ✓ Known-malicious package signatures
What it can access
- ✓ Network access No
- ✓ Filesystem access No
- ✓ Shell / process execution No
- ✓ Environment & secrets No
- ✓ Dynamic code execution No
From automated source analysis of v0.1.0. “Used” means the capability is present in the source — more access means more to trust, not that it’s unsafe.
Verified badge
Passed review? Show it. Paste this badge into your README, it links to the public security report.
Links to your public security report.[](https://agentstack.voostack.com/security/report/skill-razaib-khan-forgeweave-agent-spawner) Reliability & compatibility
Declared compatibility
Compatibility is declared by the source manifest. End-to-end runtime verification is coming, see below.
We're building live execution health for every listing: tool-call success rate, median latency, uptime, and last-checked timestamps, measured, not self-reported. It isn't live yet, so we don't show numbers we can't stand behind.
How agent discovery & health will work →About
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 |
Source & license
This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.
- Author: Razaib-khan
- Source: Razaib-khan/ForgeWeave
- License: MIT
Install and usage instructions live in the source repository linked above.
Reviews
No reviews yet, be the first.
Write a review
Versions
- v0.1.0 Imported from the upstream source.