procurement-optimizer
Pass
Audited by Gen Agent Trust Hub on May 19, 2026
Risk Level: SAFE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted JSON data from external sources to generate Markdown-formatted reports.
- Ingestion points: Scripts
spend_categorizer.py,purchasing_cycle_analyzer.py, andsupplier_consolidation.pyread user-supplied JSON files via the--inputflag. - Boundary markers: None. The scripts treat input data as raw fields (e.g., supplier names, descriptions).
- Capability inventory: Local file system read/write via
Path.read_text()andPath.write_text()in all three scripts. - Sanitization: Input strings from the JSON data are interpolated directly into Markdown strings without specific escaping, creating a surface for indirect injection if malicious data is included in the spend descriptions.
- [METADATA_POISONING]: There is a discrepancy between the author name listed in the
SKILL.mdfrontmatter ('claude-code-skills') and the actual author account ('alirezarezvani'). This is a minor metadata inconsistency.
Audit Metadata