semantic-scholar
Pass
Audited by Gen Agent Trust Hub on Sep 22, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill utilizes the
asta paperscommand suite to interface with the Semantic Scholar API, retrieving paper metadata, author information, and full-text snippets based on user-provided queries. - [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted research content from the Semantic Scholar database, creating a surface for potential indirect prompt injection.
- Ingestion points: Research abstracts, titles, and body snippets retrieved via the
asta paperstool. - Boundary markers: None explicitly defined in the output presentation guidelines for the retrieved text.
- Capability inventory: Execution is restricted to the
asta paperscommand within the Bash environment as defined in the allowed-tools configuration. - Sanitization: The skill instructions do not specify any validation or sanitization for the content returned by the external API.
Audit Metadata