genealogical-documentation
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 because it ingests untrusted user input to drive automated planning and tool usage.
- Ingestion points: User-supplied information regarding ancestor names, dates, locations, and existing records entered during the initial gathering phase (SKILL.md, Default Workflow Step 1).
- Boundary markers: The instructions lack explicit delimiters or 'ignore embedded instructions' warnings for user-provided data, which may allow malicious instructions hidden in research notes to influence the agent.
- Capability inventory: The skill utilizes the
web_searchtool and generates structured documents (plans, citations, logs) based on input (SKILL.md, Core Capabilities and Step 3). - Sanitization: There is no mention of sanitization, validation, or escaping of external content before it is interpolated into prompts or search queries.
- [SAFE]: The skill references established and well-known professional services, including FamilySearch, Ancestry, and LDSgenealogy.com, to guide the user's research strategies and citation practices.
Audit Metadata