dump
Pass
Audited by Gen Agent Trust Hub on Oct 17, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill acts as an ingestion and routing engine for unstructured user-provided content ('brainstorms' or 'brain dumps'). This creates a surface where malicious instructions embedded in the input could be promoted to execution states or influence downstream tools.
- Ingestion points: Processes 'long brainstorm or brain dump' inputs within the lifecycle of SKILL.md.
- Boundary markers: The skill does not define explicit delimiters or instructions for the agent to ignore embedded commands within the user-provided input data.
- Capability inventory: The skill description and protocol involve writing to file storage ('island/provider locator'), updating artifacts, and routing data to execution tools (
/exec) and planning tools (/propose). - Sanitization: The workflow lacks explicit sanitization, validation, or escaping mechanisms to prevent adversarial instructions in the input from influencing the AI's logic during reconciliation or routing.
Audit Metadata