anystyle-api
Pass
Audited by Gen Agent Trust Hub on Mar 31, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [SAFE]: The skill is an informative documentation resource for a well-known bibliographic tool. Its instructions and examples align with the primary purpose of parsing citation strings.
- [EXTERNAL_DOWNLOADS]: The skill interacts with the
anystyle.ioAPI and references theanystyleRuby gem repository. These external resources are legitimate dependencies for citation processing. - [COMMAND_EXECUTION]: The documentation includes shell commands for installing and executing the AnyStyle CLI (
gem install anystyle,anystyle parse). These are standard developer operations for local usage. - [PROMPT_INJECTION]: The skill has an indirect prompt injection surface because it processes untrusted, external citation data from sources like PDF documents. 1. Ingestion points: Raw citation text processed via the API and CLI in SKILL.md. 2. Boundary markers: None present. 3. Capability inventory: Network POST requests to anystyle.io and local CLI execution. 4. Sanitization: No explicit input sanitization or filtering is described in the examples.
Audit Metadata