cacos
Pass
Audited by Gen Agent Trust Hub on Sep 25, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill acts as an orchestration layer that ingests external data, creating a potential surface for indirect prompt injection.
- Ingestion points: Project requirements, book drafts, and website content provided via the
Activate CACOStask descriptions in SKILL.md. - Boundary markers: The instructions lack explicit delimiters or boundary markers to separate user-provided task data from the agent's core operational instructions.
- Capability inventory: The skill orchestrates multiple specialized agents using the
Task tooland/skillcommands, with integration points for Claude Code CLI, Agent SDK, and MCP Servers. - Sanitization: There are no observable mechanisms for sanitizing or filtering instructions that might be embedded within user-provided project data.
Audit Metadata