Agent comparison
Can Cursor make phone calls? Yes, with ClawCall
Cursor is an agent for your codebase, and it doesn’t make phone calls on its own. Add ClawCall as a remote MCP server and Cursor can place the outside call a task sometimes needs. Use Cursor alone when everything the task needs is in the code. Source: Cursor: Model Context Protocol.
Side by side
“Without ClawCall” means Cursor with nothing else connected.
| Decision factor | ClawCall + Cursor | Cursor without ClawCall |
|---|---|---|
| Primary role | Cursor supplies reasoning/orchestration; ClawCall supplies phone execution. [1][3] | Cursor edits and reasons over software. ClawCall performs bounded phone actions and returns call evidence; it does not replace Cursor’s coding models or repository tools. |
| Connection path | Customize / mcp.json using Streamable HTTP and OAuth with DCR. [1][3] | — |
| Authentication | Revocable OAuth scopes; no ClawCall API key is pasted into the host. [1][6] | — |
| Real outbound call | One user-directed call to a real US +1 number through place_call. [3][4] | Cursor can find the number and draft what to say; you place the call yourself. |
| Status model | Separate lifecycle and network outcome; the agent polls rather than guessing completion. [4] | — |
| Evidence after the call | Captured transcript evidence when present plus temporary recording access when available. [3][4] | Only the notes you take yourself. |
| Human takeover | With loop-in on, ClawCall calls the user’s verified phone, briefs them privately, and by default connects them only after they accept. [4] | — |
| Geographic scope | US +1 destinations only today. [4] | — |
| Price of phone capability | ClawCall trial and subscription terms are published on its pricing page. [5] | No ClawCall cost. |
| Action risk | Real calls can spend allowance and affect outside parties; permissions and revocation are explicit. [6][3] | Nothing is dialed, so there is no call risk. |
What changes for Cursor
Cursor connects repository context to a bounded outside action
Cursor is optimized for codebase-aware work, so the integration is most useful when a software task depends on a real phone answer. A developer can ask the agent to inspect the issue, identify the vendor or location, place one call, and attach captured transcript evidence to the investigation when present. ClawCall does not edit the code and Cursor does not become a telephony service; each product stays within the part of the workflow it is designed to handle. [1][3]
Personal, project, and team configuration change the risk
Cursor can load MCP servers from different scopes and team administrators can restrict URLs or tools. A personal server is appropriate for an individual’s calls; a project entry is easier to share but deserves repository-level trust review. The official ClawCall recommendation is to allow all thirteen tools only after that review. Leaving Cursor’s server tool allowlist empty enables the full workflow, but it should not be used to bypass a team policy or conceal a write-capable server in source control. [1][3]
Strengths and limits
What Cursor already does well
- Strong codebase context with MCP tools available directly in the editor.
- Personal, project, and team distribution options fit both individual developers and managed organizations.
ClawCall limits to know first
- ClawCall calls US numbers only, in English or 24 other languages.
- ClawCall adds phone calls. It doesn’t replace what your agent already does, like search, writing or coding.
- A call that connected isn’t proof the task got done. Check what was actually said in the transcript.
- Recording links are temporary, and every real call reaches real people.
Which setup should you choose?
Use Cursor with ClawCall when a codebase task depends on a bounded outside call. Choose personal, project, or team MCP scope deliberately; Cursor alone remains the better fit when no external phone action is needed.
Use ClawCall + Cursor when
- A codebase investigation depends on one real outside phone call.
- The team can choose personal, project, or managed MCP scope deliberately.
- Any transcript evidence actually captured should support the engineering task.
Use Cursor without ClawCall when
- You only need Cursor’s core research, reasoning, drafting, or coding capabilities.
- No external phone call is necessary or permitted by policy.
- You require non-US destinations, a language outside ClawCall’s 25, or compliance attestations ClawCall does not currently offer.
FAQ
- Does ClawCall replace Cursor?
- No. Cursor is your assistant; ClawCall is a phone it can use. It only helps when a task actually needs a call.
- What does ClawCall add to Cursor?
- Real phone calls: it gets through menus, waits on hold, brings you in when a decision is yours, and tells you what was said. It doesn’t make the model itself faster or smarter.
- Can Cursor use ClawCall without being told to?
- Yes, once it’s connected and your settings allow it. Naming ClawCall is optional.
- How do I know the call worked?
- Read what was actually said in the transcript. A call that connected doesn’t mean the task got done.
- When should I skip ClawCall?
- When no call is needed, when the number is outside the US, or when your organization doesn’t allow outside tools.
Sources
- Cursor: Model Context Protocol
- CNBC: Cursor 2026 Disruptor 50 profile
- ClawCall: Hosted MCP for AI agents
- ClawCall MCP and API documentation
- ClawCall pricing
- ClawCall security
Related pages
Give Cursor a phone line
Follow the setup guide, run the read-only check, then place one controlled call. The trial covers 30 calls or 30 minutes, whichever comes first.