schema-markup

Pass

Audited by Gen Agent Trust Hub on Apr 12, 2026

Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [PROMPT_INJECTION]: The skill's audit mode involves reading and processing HTML content from external web pages. This presents a surface for indirect prompt injection, where an attacker could embed malicious instructions in the HTML or JSON-LD data to influence the agent's analysis or subsequent actions.
  • Ingestion points: External website HTML content processed in SKILL.md (Audit Mode) and via scripts/schema_validator.py.
  • Boundary markers: Absent. The instructions do not provide specific delimiters or warnings to the agent to ignore instructions embedded within the processed data.
  • Capability inventory: Local command execution for running validation scripts.
  • Sanitization: Absent. The validation script uses standard libraries to extract data but does not perform filtering for instruction-like patterns in the extracted text.
  • [COMMAND_EXECUTION]: The skill instructs the agent to run a bundled Python script, scripts/schema_validator.py, to perform automated audits of structured data. This involves executing a local process that accepts user-supplied HTML files as input. The script itself is focused on parsing and scoring JSON-LD content and does not perform dangerous operations such as network requests or file writes.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 12, 2026, 08:16 AM
Security Audit — agent-trust-hub — schema-markup