french-language-coach
Pass
Audited by Gen Agent Trust Hub on Jun 13, 2026
Risk Level: SAFEPROMPT_INJECTIONNO_CODE
Full Analysis
- [PROMPT_INJECTION]: The skill implements a workflow in
modules/research-checklist.mdthat directs the agent to ingest and distill information from external sources (web search, Wikipedia, ArXiv). This introduces a potential surface for indirect prompt injection where malicious content from external sites could influence the agent's behavior. - Ingestion points:
modules/research-checklist.mdinstructs the agent to perform web research and use Wikipedia and ArXiv as data sources. - Boundary markers: No specific delimiters or instructions are provided to the agent to treat external content as untrusted or to ignore embedded instructions within that content.
- Capability inventory: The agent is instructed to use the results of its research to update its core guidance in
modules/core-guidance.mdand modify metadata fields such as confidence levels. - Sanitization: There are no explicit validation or sanitization steps mentioned for the external content before it is incorporated into the skill's instructions.
- [NO_CODE]: The analyzed skill consists entirely of Markdown documentation and instructional text and does not contain any executable scripts (e.g., Python, JavaScript, Shell).
Audit Metadata