stakeholder-map
Pass
Audited by Gen Agent Trust Hub on Sep 23, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection due to its processing of untrusted data from user-provided files and variables without boundary isolation.
- Ingestion points: The skill actively encourages the agent to read user-provided files like org charts, project briefs, and team rosters, and uses the
$ARGUMENTSvariable to interpolate user input directly into the context inSKILL.md. - Boundary markers: There are no explicit delimiters or instructions provided to the agent to distinguish between its primary instructions and potentially malicious commands embedded in the processed stakeholder data.
- Capability inventory: The skill requires the agent to save the generated content as a markdown document, which implies a file-write capability that could be targeted by an injection attack.
- Sanitization: The instructions do not define any validation or sanitization requirements for the content extracted from external documents before it is processed or written to disk.
Audit Metadata