frappe-agent-architect

Pass

Audited by Gen Agent Trust Hub on Jun 14, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it ingests untrusted business requirements to generate architectural roadmaps and implementation steps. Malicious instructions embedded in user-provided requirements could influence the agent to generate harmful commands or scripts.
  • Ingestion points: Business requirements, user roles, and data entity descriptions are ingested during the analysis phase (documented in SKILL.md and references/workflow.md).
  • Boundary markers: There are no specified delimiters or instructions to the agent to disregard potentially malicious commands embedded within the requirements text.
  • Capability inventory: The skill instructs the agent to generate and potentially execute implementation steps involving shell commands (e.g., bench new-app) and Python code (e.g., hooks.py, DocType controllers, and overrides).
  • Sanitization: No sanitization, validation, or escaping procedures are defined for the input requirements before they are processed into the architecture output.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 14, 2026, 09:39 AM
Security Audit — agent-trust-hub — frappe-agent-architect