oauth.start

THE RISK

High Risk

While OAuth authentication itself is a standard security practice, the tool executes an external authentication operation that creates stateful connections. This is Execute rather than Write because it triggers an external protocol flow (OAuth) whose effects go beyond simple data creation—it establishes authenticated sessions/tokens for integrations.

From the tool's definition The tool 'oauth.start' initiates an OAuth flow to authenticate and mint a connection. This triggers an external authentication process with side effects (creating a persistent connection/credential) whose outcome depends on the OAuth provider's response and…

Documented attack patterns abuse exactly the kind of access oauth.start gives an agent:

HOW TO CONTROL OAUTH.START

PolicyLayer is an MCP gateway — it sits between your AI agents and Executor, and nothing reaches the server without passing your rules. This is the rule we recommend for oauth.start:

policy.json

{
  "version": "1",
  "default": "deny",
  "tools": {
    "oauth.start": {
      "limits": [
        {
          "counter": "oauth.start_rate",
          "window": "minute",
          "max": 10,
          "scope": "grant"
        }
      ]
    }
  }
}

oauth.start stays usable, but rate-capped — a runaway agent can't fire it dozens of times a minute. Everything else on the server is denied unless you say otherwise.

Create a free account and register Executor — nothing to install.
Add this policy — paste it, or build it visually.
Point your MCP client (Claude, Cursor, anything) at your gateway URL.

RATE-LIMIT THIS TOOL →

Free to start. No card required.

EXPLORE

More Executor tools

Destructive connections.remove Remove a saved connection and its produced tools by owner, integration, and connection nam Destructive delete A destructive tool Destructive delete_titled A destructive tool with a title annotation Destructive oauth.cancel Cancel an in-flight OAuth authorization-code session by state after the user abandons the Destructive oauth.clients.remove Remove an owner-scoped OAuth client by owner and slug. Existing connections are not cascad Destructive policies.remove Remove a tool policy by id + owner. Execute connections.refresh Re-run an integration Execute probeEndpoint Probe a remote MCP endpoint before adding it. If the result requires OAuth, run the core O

All 29 Executor tools →

Execute tools on other servers

Frida Game Hacking MCP remove_breakpoint Procmon request_elevation WireMCP capture_packets Pentest Ai run_probe

Go deeper

High-risk tools in Executor → Every high-risk tool this server exposes, in one view.
All high-risk MCP tools → Every high-risk tool across the catalogue, and the policy pattern that contains them.
The MCP Attack Database → Documented attack patterns against MCP deployments — and the policies that stop them.
Rate Limiting MCP Tool Calls: A Practical Guide →
MCP Security: Why Prompt Guardrails Aren't Enough →

FAQ

What does the oauth.start tool do? +

Start OAuth through a registered client to mint a connection for an integration. It is categorised as a Execute tool in the Executor MCP Server, which means it can trigger actions or run processes. Use rate limits and argument validation.

How do I enforce a policy on oauth.start? +

Register the Executor MCP server in PolicyLayer and add a rule for oauth.start: allow, deny, rate-limit, or require approval. Point your MCP client at the PolicyLayer proxy URL and the rule is enforced on every call, before it reaches Executor. Nothing to install.

What risk level is oauth.start? +

oauth.start is a Execute tool with high risk. Execute tools should be rate-limited and have argument validation enabled.

Can I rate-limit oauth.start? +

Yes. Add a rate_limit block to the oauth.start rule in your PolicyLayer policy. For example, setting max: 10 and window: 60 limits the tool to 10 calls per minute. Rate limits are tracked per agent session and reset automatically.

How do I block oauth.start completely? +

Set action: deny in the PolicyLayer policy for oauth.start. The AI agent will receive a policy violation error and cannot call the tool. You can also include a reason field to explain why the tool is blocked.

What MCP server provides oauth.start? +

oauth.start is provided by the Executor MCP server (rhyssullivan/executor). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.

Enforce policy on every Executor tool call.

Deterministic rules across all 29 Executor tools. Per-identity grants. Full audit log. Live in minutes. Nothing to install.

GOVERN EXECUTOR →

Free to start. No card required.

29 Executor tools catalogued and risk-classified — across an index of 42,500+ MCP servers.

// WHAT OAUTH.START ON EXECUTOR DOES

// THE RISK

// HOW TO CONTROL OAUTH.START

// EXPLORE