poc-date-extractor
Pass
Audited by Gen Agent Trust Hub on Mar 25, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill processes untrusted HTML data (
html_content), creating a surface for indirect prompt injection. - Ingestion points: The skill accepts raw HTML from inspection reports as input in SKILL.md.
- Boundary markers: No explicit delimiters or instructions are provided to the agent to ignore potentially malicious instructions embedded within the processed HTML.
- Capability inventory: The skill includes logic for date parsing and string matching in Python. No dangerous capabilities such as network access, shell command execution, or dynamic code evaluation are utilized.
- Sanitization: The skill performs date normalization but lacks specific sanitization or filtering to prevent the agent from potentially obeying instructions hidden in the input HTML content.
Audit Metadata