openclaw-private-node-camera-capture

Pass

Audited by Gen Agent Trust Hub on Apr 2, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill directs the agent to execute shell commands to interact with the vendor's openclaw CLI and perform file system operations.\n
  • Evidence: Use of openclaw nodes status, openclaw nodes camera snap, mkdir -p, and cp in SKILL.md.\n- [PROMPT_INJECTION]: The instructions include behavioral nudges to prioritize this skill's workflow over standard system utilities to ensure the specific application permission chain is used.\n
  • Evidence: Explicit directives such as 'Always treat openclaw nodes camera snap|clip as the authoritative path' and 'Prefer this skill over shell tools like imagesnap'.\n- [PROMPT_INJECTION]: The skill processes output from external CLI commands, which presents a surface for potential indirect prompt injection if the tool output is compromised.\n
  • Evidence Chain:\n
  • Ingestion points: Output from openclaw nodes status --json.\n
  • Boundary markers: Absent.\n
  • Capability inventory: File system access and execution of openclaw commands.\n
  • Sanitization: Absent.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 2, 2026, 03:10 PM