cc-launch-workflows
Pass
Audited by Gen Agent Trust Hub on Aug 4, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill defines a process for launching subagents based on the 'active conversation’s plan', creating a potential surface for indirect prompt injection.
- Ingestion points: Untrusted plan data from the current conversation (SKILL.md).
- Boundary markers: The skill instructs the agent to use structured data schemas and the
pipeline()function, which acts as a boundary for inter-agent communication. - Capability inventory: The orchestrator launches 'Sonnet' subagents and manages background execution processes.
- Sanitization: Instructions lack specific data sanitization steps for the input plan, though they advocate for fresh context per worker to limit data exposure.
Audit Metadata