obs-recording

Pass

Audited by Gen Agent Trust Hub on Sep 25, 2026

Risk Level: SAFEDATA_EXFILTRATIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [DATA_EXFILTRATION]: Hardcoded local directory path exposes internal host system structure and username information.
  • Evidence: C:\Users\frank\starlight\capture\{YYYY-WW}\ in SKILL.md.
  • [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and act upon data from external sources (scripts and verbal cues) which could be manipulated to execute unintended actions within the OBS environment.
  • Ingestion points: The skill explicitly processes "scripts loaded in Speakflow", instructions from a "brief", and "verbal cues" from the performer.
  • Boundary markers: No boundary markers or instructions to ignore embedded commands are present in the skill definition.
  • Capability inventory: The skill utilizes tools to control the obs-mcp server, including starting/stopping recordings and switching scenes.
  • Sanitization: No sanitization or validation logic is specified for the external data inputs.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 25, 2026, 09:15 AM