neg-conflict-finder
Pass
Audited by Gen Agent Trust Hub on Oct 26, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill uses official Google Ads Scripting APIs to perform its documented task. The logic is transparent and focused solely on identifying keyword conflicts within the user's accounts. No malicious activity, obfuscation, or persistence mechanisms were detected.
- [DATA_EXPOSURE]: The script reads sensitive keyword data and writes it to a Google Sheet. This is an intended and documented behavior. The data transfer is restricted to the Google ecosystem.
- [INDIRECT_PROMPT_INJECTION]: The script processes keyword strings from the Google Ads environment.
- Ingestion points: Keyword text is retrieved via GAQL queries in scripts/mcc-neg-keyword-conflict.js.
- Boundary markers: None required as the data is not processed by an LLM.
- Capability inventory: No dangerous operations like subprocess execution or external web requests are present.
- Sanitization: Input strings are normalized for case and match type symbols before comparison.
Audit Metadata