cc-launch-workflows
Pass
Audited by Gen Agent Trust Hub on Sep 21, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill coordinates the execution of subagents based on an "active conversation's plan." This creates a vulnerability surface where malicious instructions embedded in a plan (such as one derived from a pull request or third-party documentation) could influence the actions of worker subagents.
- Ingestion points: The "current active conversation's plan" serves as the source for workflow generation as described in SKILL.md.
- Boundary markers: The instructions do not mandate the use of delimiters or specific warnings to ignore embedded instructions within plan content.
- Capability inventory: The orchestration involves launching subagents via
agent()andpipeline()tools that have the capability to perform code changes, refactors, and feature additions across the project. - Sanitization: The skill lacks requirements for sanitizing or filtering plan steps before they are dispatched to worker agents.
Audit Metadata