form-decompile
Pass
Audited by Gen Agent Trust Hub on Nov 2, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted XML data from external Form.xml files. This creates a potential surface where malicious instructions embedded within the XML content could influence the agent's behavior when it processes the resulting JSON output.
- Ingestion points: The skill reads file content from the path provided via the FormPath argument in SKILL.md.
- Boundary markers: The skill does not implement specific delimiters or instructions to isolate untrusted data or to warn the agent against executing embedded instructions.
- Capability inventory: The skill has access to file system read and write operations through the provided PowerShell and Python scripts to decompile forms and save output JSON and SQL files.
- Sanitization: The skill performs standard XML parsing and JSON escaping but does not specifically sanitize or filter the extracted strings for potentially malicious prompt content.
Audit Metadata