science-writing
Pass
Audited by Gen Agent Trust Hub on Feb 19, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [External Downloads] (LOW): The skill requires the
requestslibrary and interacts with the external CrossRef REST API (api.crossref.org). While CrossRef is a reputable academic service, the domain is not on the predefined whitelist. - [Command Execution] (LOW): The skill utilizes the
Bashtool to execute a local Python script for DOI validation and metadata retrieval. - [Indirect Prompt Injection] (LOW): The skill is designed to process user-provided bibliography files and manuscript sections which could be used as an attack surface for command injection or prompt manipulation.
- Ingestion points: Bibliography files (.bib, .txt) and manuscript drafts provided as user input.
- Boundary markers: Absent; there are no specific instructions or delimiters used to isolate untrusted data during processing.
- Capability inventory: The skill has access to
Bash,WebFetch,Read,Write, andEdittools. - Sanitization: There is no evidence of input validation or sanitization for strings like DOIs or filenames before they are passed to shell commands.
Audit Metadata