cookie-policy-malik-taiar
Pass
Audited by Gen Agent Trust Hub on May 15, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill utilizes the WebFetch tool to access the official CNIL website (cnil.fr) to retrieve up-to-date regulatory guidelines. This is a well-known and trusted authoritative source for data protection in France, and the operation is strictly necessary for the skill's primary function.
- [SAFE]: The drafting workflow involves analyzing user-provided URLs to identify cookies and existing policies. While this represents an indirect prompt injection surface where external web content is ingested into the agent context, the security risk is negligible as the skill lacks exploitable capabilities such as shell access, file system modification, or credential access.
- Ingestion points: The WebFetch tool reads content from the website URL provided by the user in Step 2 of the drafting workflow.
- Boundary markers: No specific delimiters or instructions to ignore embedded commands are used for the ingested web content.
- Capability inventory: Limited to Read and WebFetch tools; no dangerous capabilities like subprocess execution or file writes are present.
- Sanitization: No sanitization of external web content is mentioned.
- [SAFE]: No malicious patterns, such as obfuscated code, hardcoded credentials, persistence mechanisms, or unauthorized command execution, were detected within the skill's instructions or reference materials.
Audit Metadata