Relevant when…
Your main question is whether an agent acts within the authority your team intended.
This is editorial fit guidance based on public positioning. We have not independently tested this provider or confirmed availability.
What the provider documents
- Assessment scope
- AI agent testing · MCP assessment · Tool misuse
- Engagement model
- Assessment service
- Deliverables
- Capability and authority records, findings, evidence, remediation guidance and a regression check.
- Deployment
- Staging and test environments are supported; production access is not required for standard assessments.
- Pricing
- Not published
- Retesting
- A regression check is listed; confirm its scope.
Source for documented facts: the provider’s official offering ↗. “Not published” means not recorded in the reviewed source, not that an option is unavailable.
Before you request a proposal
- Does your system fit the standard assessment path?
- How are multi-agent and cumulative-action scenarios handled?
- What access is needed at the tool-calling boundary?
Ask the provider to put authorized systems, exclusions, evidence handling, timing and acceptance criteria in the written scope. Do not send credentials or vulnerability evidence through an initial inquiry.
Sources & verification
- Themis — official service information ↗
Supports the summary, service model, scope and recorded engagement facts above. Evidence status: provider-stated.