decision-mapping
Pass
Audited by Gen Agent Trust Hub on Jun 27, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill contains a vulnerability surface for indirect prompt injection. It instructs the agent to load the 'whole map' (a markdown file) as full context during the 'Resume' workflow, allowing content within the map to potentially influence or override the agent's instructions.
- Ingestion points: The skill reads decision map markdown files into the agent's context as part of its core logic in SKILL.md.
- Boundary markers: Absent; there are no instructions to use delimiters or to treat the ingested markdown content as untrusted data rather than instructions.
- Capability inventory: The skill is capable of invoking other high-privilege tools such as
/prototype(which likely involves code generation or execution),/grilling, and/domain-modeling, and it has permission to modify the local filesystem. - Sanitization: Absent; the skill does not perform any validation, escaping, or filtering of the markdown content before it is interpolated into the session context.
Audit Metadata