Skip to content
MCP server

Oakallow

Hosted permission checks, human approvals and audit logs for AI agent tool calls

Install
Pricing
Free trial
Platforms
API, Web
Runs
Remotely
Version
1.1.0
Listing updated
2 days ago

About Oakallow

Oakallow is an MCP server for security. It runs as a hosted remote server, so there is nothing to install locally.

Oakallow sits between an agent and the actions it wants to take. Before a tool runs, a permission check returns allow, deny or approval required. Rules can be scoped by tenant, organization, resource, tool and method, and the site describes a 12-level resolution chain evaluated at the edge on Cloudflare Workers. Each verdict comes with a single-use, HMAC-signed receipt, and every check and execution is logged with full parameters.

The MCP connector is a requester, not a decider. The README says it can list tools, check permissions, request approvals and return signed receipts, but it cannot approve or deny for a human; decisions are made on a separate dashboard or mobile app under enforced multi-factor authentication. Checking a tool that was never registered auto-creates a gated draft entry with fail-closed defaults, so an unknown tool is never trusted by default. Accounts with several organizations pass an org argument so the right rules and approvers apply.

The same governance is available through a REST API. Billing is by credit pack for approval decisions, with no subscription, and the site offers $5 in free credits to start.

Install

Run in a terminal

claude mcp add --transport http oakallow https://api.oakallow.io/mcp

Add to Settings → Connectors → Add custom connector

https://api.oakallow.io/mcp
Install in CursorOpens Cursor, which asks you to confirm. Or add it by hand:

Add to ~/.cursor/mcp.json

{
  "mcpServers": {
    "oakallow": {
      "url": "https://api.oakallow.io/mcp"
    }
  }
}
Install in VS CodeOpens VS Code, which asks you to confirm. Or add it by hand:

Add to .vscode/mcp.json

{
  "servers": {
    "oakallow": {
      "type": "http",
      "url": "https://api.oakallow.io/mcp"
    }
  }
}

Add to ~/.codeium/windsurf/mcp_config.json

{
  "mcpServers": {
    "oakallow": {
      "serverUrl": "https://api.oakallow.io/mcp"
    }
  }
}

Add to your client's MCP settings (most use this shape)

{
  "mcpServers": {
    "oakallow": {
      "type": "http",
      "url": "https://api.oakallow.io/mcp"
    }
  }
}

Tools this server provides (4)

Taken from the project documentation; the running server may differ.

list_my_tools
Lists the tools available to the signed-in user in the named org.
check_permission
Reports whether a tool call would be allowed, need approval or be blocked.
list_pending_approvals
Lists approval requests still waiting for a human decision.
check_approval_status
Polls a pending approval request by reference number.

What it does

Allow, deny or approve
Each tool call is checked against rules and returns a clear verdict.
Human-in-the-loop approvals
Risky actions create an approval request that a reviewer decides in the dashboard.
Signed decision receipts
Single-use HMAC-signed receipts prove what was decided for the exact request.
Tenant and org scoping
Tenant-scoped rules take precedence over org-wide ones.
Audit trail
Checks, approvals and executions are logged and queryable by API and dashboard.
Fail-closed unknown tools
Unregistered tools get a gated draft entry instead of being trusted.
REST and MCP
The same governance is reachable from a backend over REST or from a model over MCP.

Good for

  • Require sign-off before an agent issues a refund
  • Give each customer tenant its own tool permissions
  • Keep an audit record of what an agent did and why it was allowed
  • Poll the status of a pending approval from an agent
  • Triage newly discovered tools from the dashboard

Pricing

Free trial

$5 free credit, then prepaid credit packs; no subscription; prices per call not shown

Prices change. Check oakallow.com before you buy.

Oakallow FAQ

Can the agent approve its own requests?

No. The README says the connector cannot approve or deny; decisions happen on a separate MFA-bound dashboard.

How is it billed?

By credit packs that fund approval decisions and extended storage of audit data, with no subscription. New accounts get $5 in credits and no card is needed.

What is free?

Tool and org configuration, permission rules, the API Builder, status queries and management-key calls.

What if I have several organizations?

Pass the org argument naming the target org; if omitted when you have more than one, the call is refused.

Does check_permission create an approval?

No, it returns a read-only verdict. It may auto-create a gated draft entry for an unregistered tool.

Is Oakallow free?

Oakallow is a paid product with a free trial.

How do I add Oakallow to Claude Code?

Run this command in your terminal: claude mcp add --transport http oakallow https://api.oakallow.io/mcp

What are the best alternatives to Oakallow?

The closest alternatives to Oakallow are REA, Codebase Memory, GitHub, Serena MCP: the IDE for your agent and XERJ.

Alternatives to Oakallow

  1. REA

    MCP server

    CLI and MCP server that lets coding agents reverse engineer apps and binaries

    Runs locally
    • Open source
    • 464K GitHub stars
    • Developer tools
  2. Codebase Memory

    MCP server

    Local MCP server that builds a code knowledge graph for AI coding agents

    Runs locally
    • Open source
    • 48.7K GitHub stars
    • Knowledge and memory
  3. GitHub

    MCP server

    Official MCP server connecting AI assistants to GitHub repos, issues and workflows

    Not checked yet
    • Remote
    • Open source
    • 34.5K GitHub stars
    • Developer tools
  4. MCP toolkit giving coding agents symbol-level code retrieval and editing

    Runs locally
    • Open source
    • 31.5K GitHub stars
    • Developer tools
  5. XERJ

    MCP server

    Local search engine for AI agents with cited answers over your code and docs

    Runs locally
    • Open source
    • 19.2K GitHub stars
    • Knowledge and memory
  6. Ouroboros

    MCP server

    Spec-first layer that pins requirements and verifies results of AI coding agent work

    Runs locally
    • Open source
    • 6.4K GitHub stars
    • Developer tools