canary-watch
Pass
Audited by Gen Agent Trust Hub on Jul 29, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill has a surface for indirect prompt injection because it is designed to ingest and analyze untrusted content from external URLs.
- Ingestion points: Deployed application URLs and API endpoints provided as arguments to the canary-watch tool.
- Boundary markers: No specific delimiters or instructions to ignore embedded commands are defined for the processed web content or console errors.
- Capability inventory: The instructions describe network read operations to fetch web content and file system write access to log data to
~/.gemini/canary-watch.log. - Sanitization: The instructions do not include any steps for sanitizing, escaping, or validating the content retrieved from external sources before it is analyzed or logged.
Audit Metadata