oma-scholar

Pass

Audited by Gen Agent Trust Hub on Jun 13, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTIONCREDENTIALS_UNSAFE
Full Analysis
  • [EXTERNAL_DOWNLOADS]: Fetches academic paper metadata and structured sidecars from public research services including Knows Academy and OpenAlex.
  • [COMMAND_EXECUTION]: Executes the oma scholar CLI for searching academic databases and validating locally generated YAML sidecar files.
  • [CREDENTIALS_UNSAFE]: Includes instructions for the management of API keys for OpenAlex, which advise users to store credentials in local environment variables or .env files rather than hardcoding them.
  • [PROMPT_INJECTION]: The skill processes untrusted data from research papers and remote sidecars; it provides validation via a linting tool and specific anti-fabrication instructions to mitigate the risks of indirect prompt injection.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 13, 2026, 12:18 PM
Security Audit — agent-trust-hub — oma-scholar