frappe-syntax-hooks-events

Pass

Audited by Gen Agent Trust Hub on Jun 25, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [SAFE]: The skill provides instructional content and code examples for the Frappe framework. The content is educational in nature and aligns with established development practices for that platform.\n- [PROMPT_INJECTION]: The skill describes mechanisms for an agent to process document data via event hooks, which represents an indirect prompt injection surface.\n
  • Ingestion points: Document object fields processed in event handlers like validate or on_submit (SKILL.md, references/examples.md).\n
  • Boundary markers: None present in the provided code snippets.\n
  • Capability inventory: Database operations (doc.db_set, doc.insert), background task processing (frappe.enqueue), and automated email dispatch (frappe.sendmail) (references/examples.md).\n
  • Sanitization: Not explicitly mentioned in the instructional snippets, which focus on framework logic.\n- [SAFE]: A metadata discrepancy was noted where the author in the frontmatter ('OpenAEC-Foundation') does not match the system-identified author ('Impertio-Studio'). This appears to be a documentation oversight rather than a malicious attempt at deception.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 25, 2026, 11:33 AM
Security Audit — agent-trust-hub — frappe-syntax-hooks-events