input-validation

Pass

Audited by Gen Agent Trust Hub on Mar 5, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection when processing untrusted construction data.
  • Ingestion points: The instructions.md file specifies that the agent should accept data from file paths or direct input in formats like CSV, Excel, and JSON.
  • Boundary markers: The skill does not implement delimiters or 'ignore' instructions to distinguish between data and potential malicious commands within the inputs.
  • Capability inventory: The skill has filesystem permissions and uses Python logic to process and validate data, which could be leveraged if an injection is successful.
  • Sanitization: Input validation is limited to domain-specific types (e.g., date formats, CSI code regex) and does not filter for instructional content in text fields.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 5, 2026, 04:29 AM