competitors
Pass
Audited by Gen Agent Trust Hub on Sep 7, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill instructions direct the agent to automatically read local project files, such as
.agents/product-marketing.md, to gather context for generating competitor comparison pages. This ingestion process lacks security controls to prevent malicious data within those files from influencing the agent's behavior. - Ingestion points:
.agents/product-marketing.md,.claude/product-marketing.md, andproduct-marketing-context.md(referenced inSKILL.md). - Boundary markers: Absent; there are no delimiters or specific instructions to treat the external file content as untrusted.
- Capability inventory: None identified within the skill's own files; the skill utilizes the agent's base text generation capabilities.
- Sanitization: Absent; the skill does not include any logic or instructions to filter or sanitize the content ingested from external files.
Audit Metadata