market
Pass
Audited by Gen Agent Trust Hub on Oct 11, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests venture-specific data from 'innov.local.md' to customize its market analysis reports. This configuration file represents an ingestion surface for indirect prompt injection if its contents are derived from untrusted external sources.
- Ingestion points: 'innov.local.md' (referenced in the CONTEXT LOADING section of SKILL.md).
- Boundary markers: Absent; the skill extracts fields like 'problem_statement' and 'target_customer' directly into the analysis prompts without specific delimiters or isolation instructions.
- Capability inventory: The skill's capabilities are limited to generating text analysis. It does not utilize any tools for file system writing, network requests, or shell command execution.
- Sanitization: Absent; the instructions do not include procedures for sanitizing or escaping the content read from the venture context file, relying instead on the model's baseline safety filters.
Audit Metadata