scholar-search
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill fetches paper metadata from api.openalex.org, which is a recognized scientific data catalog. The communication uses standard libraries and adheres to the service's API guidelines.\n- [COMMAND_EXECUTION]: The skill includes a Python search script intended to be run by the agent to find papers based on user-provided criteria.\n- [INDIRECT_PROMPT_INJECTION]: The skill represents an attack surface for indirect prompt injection as it processes data from an external API.\n
- Ingestion points:
scripts/scholar_search.pyretrieves titles and abstracts from OpenAlex API responses.\n - Boundary markers: The skill does not currently use specific delimiters to wrap the retrieved external text.\n
- Capability inventory: The script's capabilities are limited to network GET requests to api.openalex.org and file system writes to a specific cache directory (
~/.cache/ai-skill-scholar/) for rate-limiting.\n - Sanitization: Data is parsed as JSON and reconstructed as plain text without any dynamic execution or high-privilege operations.
Audit Metadata