Delegates arbitrary instructions to sub-agents,
Claims to do
Dispatching Parallel Agents: You delegate tasks to specialized agents with isolated context. By precisely crafting their instructions and context, you ensure they stay focused and succeed at their task. They should never inherit your session's context or history — you construct exactly what they need. This also preserves your own context for coordination work.
Actually does
This skill describes a conceptual pattern for an AI agent to dispatch multiple sub-agents to work on independent tasks concurrently. It instructs the main agent to create isolated tasks (e.g., `Task("Fix ...")`) for sub-agents, providing them with specific scope, goals, and constraints, and then integrating their results. It does not directly execute external tools, access specific data paths, or contact URLs itself, but rather outlines a strategy for *how* an AI system should manage parallel work.
/plugin marketplace add obra/superpowers/plugin install dispatching-parallel-agents@obra/superpowersgemini extensions install https://github.com/obra/superpowers.git --consentnpx skills add https://github.com/obra/superpowersThe skill describes a pattern for dispatching sub-agents using a `Task()` function. This mechanism allows the parent agent to delegate arbitrary instructions to sub-agents, which can then perform actions like reading/writing files and modifying code. While the examples are benign, this capability could be abused to execute arbitrary commands, exfiltrate data, or perform other malicious actions if the parent agent is compromised or given malicious input.
```typescript
Task("Fix agent-tool-abort.test.ts failures")
// ...
Fix the 3 failing tests in src/agents/agent-tool-abort.test.ts:
...
Your task:
1. Read the test file and understand what each test verifies
2. Identify root cause - timing issues or actual bugs?
3. Fix by:
- Replacing arbitrary timeouts with event-based waiting
- Fixing bugs in abort implementation if found
- Adjusting test expectations if testing changed behavior
```[](https://mondoo.com/ai-agent-security/skills/github/obra/superpowers/dispatching-parallel-agents)<a href="https://mondoo.com/ai-agent-security/skills/github/obra/superpowers/dispatching-parallel-agents"><img src="https://mondoo.com/ai-agent-security/api/badge/github/obra/superpowers/dispatching-parallel-agents.svg" alt="Mondoo Skill Check" /></a>https://mondoo.com/ai-agent-security/api/badge/github/obra/superpowers/dispatching-parallel-agents.svgSkills can read files, run commands, and access credentials. Mondoo helps organizations manage the security risks of AI agent skills across their entire fleet.