Silverfort: governance profile
AI Agent Security released 18 June 2025; Google Cloud Agent Gateway integration announced 12 October 2026. Rank 2, total 8.35.
In short: Silverfort ranks second at 8.35 out of 10. It discovers sanctioned and rogue agents across identity providers, clouds and SaaS. It binds each one to a human owner and requires approval before sensitive tool actions. Its policy language is not published and a revocation workflow is not described, which holds back policy and lifecycle.
How does Silverfort score in each area?
| Area | Weight | Score | Reason | Source |
|---|---|---|---|---|
| A1 Inventory and discovery | 15 | 9 | It discovers sanctioned and rogue agents across identity providers, clouds and SaaS through read-only APIs. | Source: Silverfort: AI Agent Security, read |
| A2 Ownership and accountability | 20 | 9 | It binds each agent to a human owner. | Source: Silverfort: AI Agent Security, read |
| A3 Policy authoring and enforcement | 20 | 8 | Its MCP Gateway evaluates each call against authorization planes and scopes with approve, block or step-up verdicts, but the policy language is not published. | Source: Silverfort: AI Agent Security; Silverfort unveils AI Agent Security, read |
| A4 Human approval and escalation | 15 | 9 | It requires approvals before sensitive tool actions. | Source: Silverfort: AI Agent Security, read |
| A5 Audit evidence | 15 | 9 | It keeps an immutable audit trail that links each action to the person who started it. | Source: Silverfort: AI Agent Security, read |
| A6 Lifecycle and revocation | 10 | 6 | It removes standing privileges, but a revocation workflow is not described. | Source: Silverfort: AI Agent Security, read |
| A7 Buying clarity | 5 | 6 | Its packages are quote-based. | Source: Silverfort pricing, read |
Which of the 24 controls does Silverfort document?
Not found means we did not find the control described in the sources listed. It does not mean the product lacks it.
- GOV-01Agent inventory: documentedSource: Silverfort: AI Agent Security, read
- GOV-02Named accountable owner: documentedSource: Silverfort: AI Agent Security, read
- GOV-03Documented purpose: not found in public materialChecked: Silverfort: AI Agent Security; Silverfort unveils AI Agent Security; Silverfort and Google Cloud Agent Gateway; Silverfort pricing, read
- GOV-04Least privilege at creation: not found in public materialChecked: Silverfort: AI Agent Security; Silverfort unveils AI Agent Security; Silverfort and Google Cloud Agent Gateway; Silverfort pricing, read
- GOV-05Short-lived credentials: partly documentedSource: Silverfort: AI Agent Security, read
- GOV-06Per-action policy: documentedSource: Silverfort: AI Agent Security, read
- GOV-07Human approval for high-risk actions: documentedSource: Silverfort: AI Agent Security, read
- GOV-08No self-approval: not found in public materialChecked: Silverfort: AI Agent Security; Silverfort unveils AI Agent Security; Silverfort and Google Cloud Agent Gateway; Silverfort pricing, read
- GOV-09Policy testing before go-live: not found in public materialChecked: Silverfort: AI Agent Security; Silverfort unveils AI Agent Security; Silverfort and Google Cloud Agent Gateway; Silverfort pricing, read
- GOV-10Tool allow-listing and hiding: not found in public materialChecked: Silverfort: AI Agent Security; Silverfort unveils AI Agent Security; Silverfort and Google Cloud Agent Gateway; Silverfort pricing, read
- GOV-11Revocation and kill switch: not found in public materialChecked: Silverfort: AI Agent Security; Silverfort unveils AI Agent Security; Silverfort and Google Cloud Agent Gateway; Silverfort pricing, read
- GOV-12Decommission at task end: not found in public materialChecked: Silverfort: AI Agent Security; Silverfort unveils AI Agent Security; Silverfort and Google Cloud Agent Gateway; Silverfort pricing, read
- GOV-13Per-action audit log tied to a person: documentedSource: Silverfort: AI Agent Security, read
- GOV-14SIEM export: not found in public materialChecked: Silverfort: AI Agent Security; Silverfort unveils AI Agent Security; Silverfort and Google Cloud Agent Gateway; Silverfort pricing, read
- GOV-15Periodic access review: not found in public materialChecked: Silverfort: AI Agent Security; Silverfort unveils AI Agent Security; Silverfort and Google Cloud Agent Gateway; Silverfort pricing, read
- GOV-16Shadow agent detection: documentedSource: Silverfort: AI Agent Security, read
- GOV-17MCP server registry: partly documentedSource: Silverfort: AI Agent Security, read
- GOV-18Delegation bounded by the person's permissions: not found in public materialChecked: Silverfort: AI Agent Security; Silverfort unveils AI Agent Security; Silverfort and Google Cloud Agent Gateway; Silverfort pricing, read
- GOV-19Agent incident response: not found in public materialChecked: Silverfort: AI Agent Security; Silverfort unveils AI Agent Security; Silverfort and Google Cloud Agent Gateway; Silverfort pricing, read
- GOV-20Change management for agent permissions: not found in public materialChecked: Silverfort: AI Agent Security; Silverfort unveils AI Agent Security; Silverfort and Google Cloud Agent Gateway; Silverfort pricing, read
- GOV-21Third-party agent onboarding: not found in public materialChecked: Silverfort: AI Agent Security; Silverfort unveils AI Agent Security; Silverfort and Google Cloud Agent Gateway; Silverfort pricing, read
- GOV-22Data access boundaries: not found in public materialChecked: Silverfort: AI Agent Security; Silverfort unveils AI Agent Security; Silverfort and Google Cloud Agent Gateway; Silverfort pricing, read
- GOV-23Evidence retention: not found in public materialChecked: Silverfort: AI Agent Security; Silverfort unveils AI Agent Security; Silverfort and Google Cloud Agent Gateway; Silverfort pricing, read
- GOV-24Board reporting: not found in public materialChecked: Silverfort: AI Agent Security; Silverfort unveils AI Agent Security; Silverfort and Google Cloud Agent Gateway; Silverfort pricing, read
What does Silverfort cost?
Packages are quote-based. No per-agent price is published.
Does Silverfort work alongside your identity provider?
Yes in practice. Silverfort discovers agents across identity providers such as Entra ID and Okta through read-only APIs, so it reads from the provider you run.
A governance layer runs alongside your identity provider, never instead of it.
How does Silverfort rank for specific buyer needs?
- Second at 8.50Best for regulated industries
- Second at 8.50Best for human oversight
- Shares second at 8.60Best for shadow agent inventory
- Second at 8.55Best for audit evidence
- Second at 7.75Best for a small team's fast start
- Shares second at 7.65Best for lifecycle and revocation
Head-to-head comparisons
Frequently asked questions
- 6.1 What is Silverfort best at?
- Its highest area scores are inventory, ownership, approval and audit evidence, at 9 out of 10. It discovers sanctioned and rogue agents across identity providers, clouds and SaaS through read-only APIs.
- 6.2 Where are the gaps in Silverfort?
- Its lowest area is lifecycle, at 6 out of 10. It removes standing privileges, but a revocation workflow is not described. Controls not found in its public material: 16 of 24.
- 6.3 How is Silverfort priced?
- Packages are quote-based. No per-agent price is published.
Reviewed by The Register desk