skills/isaaccorley/skills/bib-audit/Gen Agent Trust Hub

bib-audit

Pass

Audited by Gen Agent Trust Hub on Jul 25, 2026

Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill documentation and SKILL.md body suggest using common command-line utilities for PDF text extraction and reference auditing (e.g., pdftotext, python3 scripts/validate_refs.py). These commands are standard for the tool's primary purpose of academic reference auditing and do not involve privilege escalation, persistence, or dangerous shell injection patterns.
  • [EXTERNAL_DOWNLOADS]: The scripts interact with well-known and trusted academic metadata APIs, including Crossref (api.crossref.org), arXiv (export.arxiv.org), DataCite (api.datacite.org), and Semantic Scholar (api.semanticscholar.org). These operations are documented neutrally as they are central to the skill's function of verifying bibliographic data against official registries.
  • [CREDENTIALS_UNSAFE]: The skill implements safe secret management practices. It specifically instructs users to provide a Semantic Scholar API key via an environment variable (S2_API_KEY) rather than command-line flags, preventing the key from being stored in shell history. No hardcoded credentials or private keys were found in the source code.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 25, 2026, 07:13 PM
Security Audit — agent-trust-hub — bib-audit