strategy-builder
Pass
Audited by Gen Agent Trust Hub on Nov 5, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes data from external files, including a market analysis report and a marketing strategy template. This creates a standard surface for indirect prompt injection where instructions embedded in those files could theoretically influence the agent.
- Ingestion points: The skill reads from a path defined by the
MARKET_ANALYSIS_PATHvariable and the.claude/templates/marketing-strategy-template.mdfile. - Boundary markers: The skill provides instructions to "Extract only what the evidence provenance supports," but it lacks explicit structural delimiters to isolate the ingested content from the rest of the prompt.
- Capability inventory: The skill performs file reading and file writing within the local project directory.
- Sanitization: No specific sanitization or filtering logic is defined for the content read from the source files.
- [SAFE]: The skill operates entirely within the local project environment, focusing on documentation and strategic planning. No network exfiltration, hardcoded credentials, or obfuscated payloads were detected.
Audit Metadata