academic-verify

Pass

Audited by Gen Agent Trust Hub on May 13, 2026

Risk Level: SAFEPROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [PROMPT_INJECTION]: The skill features an indirect prompt injection surface as it processes external research claims and web content to generate its reports.
  • Ingestion points: Untrusted data enters the workflow via research claims in Step 1 and external content fetched from the web in Step 3 of SKILL.md.
  • Boundary markers: The documentation does not specify the use of delimiters or instructions to ignore embedded commands within the processed data.
  • Capability inventory: The skill has the authority to write markdown pages to the concepts/ directory, creating a persistent record of processed data.
  • Sanitization: There is no description of sanitization or validation processes for the ingested external information before it is formatted into the final verdict.
  • [EXTERNAL_DOWNLOADS]: The skill refers to established academic databases such as Retraction Watch, PubPeer, and OSF. These references are appropriate for the skill's purpose and target reputable research services.
Audit Metadata
Risk Level
SAFE
Analyzed
May 13, 2026, 03:26 PM
Security Audit — agent-trust-hub — academic-verify