yc-4-planning-an-mvp

Pass

Audited by Gen Agent Trust Hub on Sep 22, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill includes an interactive self-assessment component that ingests untrusted user input and persists the synthesized results to the local file system.\n
  • Ingestion points: User responses collected via the AskUserQuestion tool as part of the module-4-workbook logic in SKILL.md.\n
  • Boundary markers: The skill does not implement specific delimiters or 'ignore instructions' warnings to prevent the model from obeying instructions potentially embedded in user responses.\n
  • Capability inventory: The skill has the capability to write and save markdown files to the knowledge/ directory based on processed data.\n
  • Sanitization: There is no evidence of explicit sanitization, filtering, or validation of user-supplied text before it is interpolated into the final stored assessment documents.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 22, 2026, 02:37 AM
Security Audit — agent-trust-hub — yc-4-planning-an-mvp