vss-search-index

Pass

Audited by Gen Agent Trust Hub on Aug 12, 2026

Risk Level: SAFE
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill includes a bootstrap script (scripts/vss-bootstrap.sh) that ensures the VSS application source is available. It follows a secure detection logic: first, it attempts to find an existing local checkout of the VSS repository in the current workspace; if not found, it performs a sparse, shallow clone of the official VSS project from https://github.com/open-edge-platform/edge-ai-libraries.git. This behavior is limited to fetching official dependencies and is documented as a necessary setup step.
  • [COMMAND_EXECUTION]: The skill provides numerous curl commands to interact with the VSS Manager API. These commands are intended to be executed by the agent to upload videos (POST /manager/videos), manage search indices, and perform natural language queries. The usage is standard for an operational API-driven skill and targets a configurable $HOST variable (defaulting to localhost).
  • [DATA_EXFILTRATION]: While the skill involves uploading video files and querying metadata, all data remains within the user's controlled VSS deployment environment. The network operations are localized to the defined VSS manager host and do not target unauthorized third-party domains.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 12, 2026, 12:12 AM
Security Audit — agent-trust-hub — vss-search-index