camsnap
Pass
Audited by Gen Agent Trust Hub on Apr 11, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill provides an interface for the
camsnapandffmpegbinaries. Thewatchcommand supports an--actionflag which executes shell commands in response to motion detection events.\n- [PROMPT_INJECTION]: The skill contains a surface for indirect prompt injection.\n - Ingestion points: The
discovercommand retrieves and processes unauthenticated metadata from devices on the local network.\n - Boundary markers: No delimiters or instructions are used to distinguish untrusted network data from agent logic.\n
- Capability inventory: The skill can execute binary commands and trigger shell actions.\n
- Sanitization: There is no evidence of validation or sanitization of device metadata before it is presented to the agent context.\n- [EXTERNAL_DOWNLOADS]: The skill references the vendor's official homepage (camsnap.ai) and requires the
camsnapandffmpegbinaries to be present in the system environment.
Audit Metadata