Skovos Governance
Registry, scoped permissions and one-step recall for AI agents in health systems
- Pricing
- Contact sales
- Platforms
- Web, API
- Runs
- Remotely
- Version
- 1.0.0
- Listing updated
- 2 days ago
About Skovos Governance
Skovos Governance is an MCP server for security. It runs as a hosted remote server, so there is nothing to install locally.
Skovos approaches agent governance from the hospital side: registration, permissions and the stop control sit with the health system rather than with each AI vendor. Every agent gets an identity, an accountable owner, a role and explicit action-and-resource scopes such as read:vitals or write:drafts. Before an agent reads a chart or writes an order, the action is checked against those scopes; anything outside is denied and recorded.
A recalled agent is denied on every later request, and the recall logs who made it and why. Decisions go into a hash-chained log the organisation keeps, and signed evidence exports use keyed commitments rather than patient data or agent names. Audit records that look like patient identifiers are rejected, and a model-armor feature detects and redacts identifiers such as SSN, MRN, email, phone and dates in agent inputs and outputs.
The MCP connector lets governance leads ask in plain language which agents touched vitals this week or stop a named agent. Skovos describes itself as HIPAA-aligned with no PHI required for core use, and offers a scoped, paid engagement that starts with one workflow.
Install
Run in a terminal
claude mcp add --transport http governance https://mcp.skovos.ai/mcpAdd to Settings → Connectors → Add custom connector
https://mcp.skovos.ai/mcpAdd to ~/.cursor/mcp.json
{
"mcpServers": {
"governance": {
"url": "https://mcp.skovos.ai/mcp"
}
}
}Add to .vscode/mcp.json
{
"servers": {
"governance": {
"type": "http",
"url": "https://mcp.skovos.ai/mcp"
}
}
}Add to ~/.codeium/windsurf/mcp_config.json
{
"mcpServers": {
"governance": {
"serverUrl": "https://mcp.skovos.ai/mcp"
}
}
}Add to your client's MCP settings (most use this shape)
{
"mcpServers": {
"governance": {
"type": "http",
"url": "https://mcp.skovos.ai/mcp"
}
}
}What it does
- Agent registry
- Identity, owner, role and scopes for every connected agent.
- Scope enforcement
- Out-of-scope actions are denied and recorded before they run.
- One-step recall
- Stop an agent without waiting on the vendor; later requests are denied.
- Hash-chained audit log
- Tamper-evident record of registrations, decisions and recalls.
- Signed evidence export
- Proof of controls that ran, without patient data or agent names.
- PHI guardrails
- Detects and redacts common identifiers in agent inputs and outputs.
Good for
- Ask which agents accessed vitals this week
- Stop a misbehaving clinical agent
- Give a governance committee an overview of denied actions
- Prepare verifiable evidence for auditors
Pricing
Contact sales
Scoped paid engagement; fee agreed per workflow
Prices change. Check skovos.ai before you buy.
Skovos Governance FAQ
Does it need patient data?
Skovos says no PHI is required for core use, and patient identifiers are rejected in audit records.
How does a paid engagement start?
With a scoped workflow; Skovos aims to deliver an access map and review packet within two business days of agreed inputs, as a target rather than a guarantee.
Can an agent be stopped?
Yes, recall is a single step and every later request from that agent is denied.
What do scopes look like?
Examples are read:vitals, write:orders and write:drafts.
How much does Skovos Governance cost?
Skovos Governance does not publish prices. Contact the company for a quote.
How do I add Skovos Governance to Claude Code?
Run this command in your terminal: claude mcp add --transport http governance https://mcp.skovos.ai/mcp
What are the best alternatives to Skovos Governance?
The closest alternatives to Skovos Governance are REA, Tooluniverse, Oya Browser, Agent Roast and Midplane.
Alternatives to Skovos Governance

REA
MCP serverCLI and MCP server that lets coding agents reverse engineer apps and binaries
Runs locally- Open source
- 464K GitHub stars
- Developer tools

Tooluniverse
MCP serverMCP server giving AI agents access to thousands of scientific research tools
Runs locally- Open source
- 1.8K GitHub stars
- Health and science

Oya Browser
MCP serverHIPAA-oriented agentic browser with replayable playbooks, reachable over MCP
- 32 tools
- Remote
- Freemium
- 349 GitHub stars
- Browser automation
Agent Roast
MCP serverOffline review that flags agent changes which weaken tests, assertions or CI
Runs locally- Open source
- 4 Hacker News points
- Developer tools

Midplane
MCP serverPostgres gateway for AI agents with table grants, column masking and approval for writes
Runs locally- Open source
- 5 GitHub stars
- Databases

Leash
MCP serverReview what your AI touched, approve risky actions and undo supported changes
- 1 tool
- Remote
- Free
- 1 GitHub star
- Security