ai-sdr
Pass
Audited by Gen Agent Trust Hub on Sep 14, 2026
Risk Level: SAFENO_CODEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill is entirely composed of markdown files providing go-to-market strategy, platform comparisons, and implementation frameworks. No automation scripts, shell commands, or configuration files were detected.\n- [INDIRECT_PROMPT_INJECTION]: The architecture described in the skill involves processing untrusted external data, such as email replies and social media signals, for classification and automated response generation. This represents a theoretical ingestion surface for indirect prompt injection in systems implemented according to these designs.\n
- Ingestion points: Processing prospect email replies and web-scraped social media signals (SKILL.md, implementation-guide.md).\n
- Boundary markers: The instructional text does not specify the use of delimiters or boundary markers for external data.\n
- Capability inventory: No implementation code or executable tools are provided within the skill.\n
- Sanitization: The documentation does not discuss sanitization or validation strategies for incoming prospect data.
Audit Metadata