xy-atomize
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill possesses an indirect prompt injection surface. 1. Ingestion points: Processes user documents from local directories (e.g., ~/Documents/). 2. Boundary markers: Explicit JSON schema and atomic extraction rules are defined in SKILL.md. 3. Capability inventory: Writes JSONL files to local directories and executes a local validation script. 4. Sanitization: Uses validate_batch.py to enforce schema, length limits, and a name blocklist.
- [COMMAND_EXECUTION]: The skill runs a local validation tool (scripts/validate_batch.py) to audit the processed content. This is a restricted execution path intended for data integrity verification.
Audit Metadata