camsnap

Warn

Audited by Gen Agent Trust Hub on Mar 30, 2026

Risk Level: MEDIUMEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill configuration (YAML frontmatter) specifies an installation step that fetches a binary from a third-party Homebrew tap (steipete/tap/camsnap). While the installation mechanism is standard, the source repository is managed by an individual developer outside of the recognized trusted organizations list.
  • [COMMAND_EXECUTION]: The camsnap watch command utilizes an --action parameter that executes arbitrary shell strings when motion events are triggered. This provides a mechanism for local command execution depending on how the tool handles the action string.
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection via the motion detection workflow.
  • Ingestion points: External metadata and event triggers from RTSP/ONVIF cameras processed by the watch command.
  • Boundary markers: None identified; the --action parameter accepts a raw string without specified delimiters or instruction-ignore warnings.
  • Capability inventory: The skill uses the camsnap binary which has the capability to execute shell commands and write files (--out).
  • Sanitization: No input sanitization or validation is described for the data being passed into the --action shell command string.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Mar 30, 2026, 01:08 AM
Security Audit — agent-trust-hub — camsnap