halo-asset-context
Pass
Audited by Gen Agent Trust Hub on Oct 2, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill tools are strictly read-only, allowing the agent to gather context about Halo assets and tickets without the ability to modify any records.
- [SAFE]: Credential management follows best practices by using environment variables (HALO_CLIENT_ID, HALO_CLIENT_SECRET) rather than hardcoded secrets.
- [INDIRECT_PROMPT_INJECTION]: The skill processes data from the Halo API, which serves as a potential surface for indirect prompt injection.
- Ingestion points: External asset details, device fields, and ticket history are ingested into the agent context via tools like
halo_get_assetandhalo_get_asset_tickets. - Boundary markers: The skill does not explicitly define delimiters or instructions to ignore potential commands embedded in the API data.
- Capability inventory: The skill is limited to read-only capabilities, significantly reducing the impact of any successful injection from external data.
- Sanitization: No specific sanitization or filtering of the Halo API responses is documented.
Audit Metadata