twingate-identity
Pass
Audited by Gen Agent Trust Hub on Sep 9, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill instructions define a specialized role for identity and access management within Twingate. The instructions focus on directing the agent to use a local corpus of reference files to provide accurate configuration guidance.
- [COMMAND_EXECUTION]: The
SKILL.mdfile provides examples ofgrepcommands (e.g.,grep -ril "fleetdm" references/) for the agent to use when searching its internal documentation. This is a standard and safe use of local tools for context retrieval within the agent's operating environment. - [EXTERNAL_DOWNLOADS]: Several reference files (e.g.,
gh-twingate-solutions-twingate-mdm-connector.md,gh-twingate-labs-tg-group-profile-manager.md) describe external automation tools, middleware, and reference implementations hosted on GitHub and GHCR. These are documented neutrally as resources for the user and originate from vendor-related organizations (twingate-solutions,twingate-labs). - [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest user keywords for searching the
references/directory. While this involves processing untrusted user input, the capability is restricted to local documentation lookups and does not expose high-risk actions. The documentation itself is static and provided by the skill author.
Audit Metadata