alterlab-research-workflows
Pass
Audited by Gen Agent Trust Hub on Sep 25, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill interacts with well-known and trusted scholarly database APIs, including OpenAlex, Crossref, PubMed, and arXiv, to verify reference metadata and retrieve academic records. These interactions are standard for academic research tools and are performed in accordance with the services' polite usage guidelines.
- [COMMAND_EXECUTION]: The workflows utilize the Bash tool to execute specific Python verification scripts (verify_citations.py and claim_faithfulness.py) from the vendor's core academic library. This is a standard and expected implementation for performing automated integrity audits on research manuscripts.
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external data such as research manuscripts, bibliography exports, and peer review comments, which constitutes a potential attack surface. However, the system is designed with multiple safeguards, including the use of independent blinded agents who do not share context, and strict verification protocols where integrity flags must be confirmed by two separate retrieval strategies before being reported.
Audit Metadata