master-nagarjuna

Pass

Audited by Gen Agent Trust Hub on Nov 7, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSDATA_EXFILTRATIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill is configured to fetch scripture content and search results from the fojin.app API when local resources are insufficient. This is a functional requirement for providing verified Buddhist citations.\n- [DATA_EXFILTRATION]: User queries are passed to the fojin.app search API to retrieve relevant religious texts. While this involves sending user input to an external service, it is the primary mechanism for the skill's search functionality.\n- [INDIRECT_PROMPT_INJECTION]: Because the skill processes content from an external API, there is a theoretical risk of indirect prompt injection. The skill mitigates this through an 'assume-malicious' posture for API data, utilizing explicit boundary markers (<<<FOJIN_DATA>>>) and strict instructions to treat retrieved content as non-executable text strings.
Audit Metadata
Risk Level
SAFE
Analyzed
Nov 7, 2026, 01:52 PM