Enables AI agents to make outbound US phone calls, navigate menus, wait on hold, and retrieve call transcripts.
Also known as ClawCall-Dev/clawcall, @clawcall/mcp
- Stars
- 10
- Tools
- —
- Downloads, 30 days
- 195
- Latest version
- 0.2.0
- Last commit
- Oct 13, 2026
- License
- MIT
What is ClawCall MCP?
ClawCall enables AI agents to make outbound US phone calls, navigate menus, wait on hold, and retrieve call transcripts.
Summary generated from the project's README
About ClawCall MCP server
Clawcall provides a way for AI agents to initiate outbound phone calls within the United States. This stdio MCP server acts as a connectivity layer for the ClawCall REST API, allowing hosts like Claude Desktop, Cursor, or Windsurf to interact with the phone network. It is designed for developers who want to integrate voice communication into their agentic workflows, enabling tasks such as navigating phone menus or waiting on hold.
Users can set up the server using npx without needing an initial API key, as the system self-provisions a proto-key during the first place_call command. This key and the user phone number are stored in a local configuration file to maintain a consistent identity, trial quota, and call history. While the server supports an optional CLAWCALL_API_KEY environment variable for overrides, it primarily relies on this local persistence to manage account state.
The server offers different interaction patterns through specific tools. The place_call_and_wait tool provides a shortcut for single calls by polling until the lifecycle is finalized, up to a 240 second ceiling. For more complex tasks like information gathering campaigns, developers can use the non-blocking place_call and get_call tools to manage multiple parallel calls.
What you can do with it
- 1Execute single outbound callsAn AI agent can place a single phone call and wait for the result to be finalized before proceeding.
- 2Run information gathering campaignsA developer can instruct an agent to perform multiple parallel calls to collect quotes or check inventory.
- 3Analyze call outcomes via transcriptsAfter a call ends, the agent can retrieve the text transcript to evaluate if the specific task was successful.
Written from the project's README. Every use case is checked against a quote from it.
How to install ClawCall MCP
Formats verified 10 Nov 2026Add to CursorAdd to GooseInstall ClawCall MCP in Claude Code
$ claude mcp add --env 'CLAWCALL_API_KEY=<YOUR_CLAWCALL_API_KEY>' clawcall -- npx -y @clawcall/mcpInstall ClawCall MCP in Cursor
~/.cursor/mcp.json
{
"mcpServers": {
"clawcall": {
"command": "npx",
"args": [
"-y",
"@clawcall/mcp"
],
"env": {
"CLAWCALL_API_KEY": "<YOUR_CLAWCALL_API_KEY>"
}
}
}
}Install ClawCall MCP in VS Code
.vscode/mcp.json
{
"servers": {
"clawcall": {
"type": "stdio",
"command": "npx",
"args": [
"-y",
"@clawcall/mcp"
],
"env": {
"CLAWCALL_API_KEY": "<YOUR_CLAWCALL_API_KEY>"
}
}
}
}Install ClawCall MCP in VS Code CLI
$ code --add-mcp '{"name":"clawcall","type":"stdio","command":"npx","args":["-y","@clawcall/mcp"],"env":{"CLAWCALL_API_KEY":"<YOUR_CLAWCALL_API_KEY>"}}'Install ClawCall MCP in Claude Desktop
claude_desktop_config.json
{
"mcpServers": {
"clawcall": {
"command": "npx",
"args": [
"-y",
"@clawcall/mcp"
],
"env": {
"CLAWCALL_API_KEY": "<YOUR_CLAWCALL_API_KEY>"
}
}
}
}Install ClawCall MCP in Codex
$ codex mcp add clawcall --env 'CLAWCALL_API_KEY=<YOUR_CLAWCALL_API_KEY>' -- npx -y @clawcall/mcpInstall ClawCall MCP in Codex config.toml
~/.codex/config.toml
[mcp_servers.clawcall] command = "npx" args = ["-y", "@clawcall/mcp"] [mcp_servers.clawcall.env] CLAWCALL_API_KEY = "<YOUR_CLAWCALL_API_KEY>"
Install ClawCall MCP in Gemini CLI
$ gemini mcp add -s user -e 'CLAWCALL_API_KEY=<YOUR_CLAWCALL_API_KEY>' clawcall npx -- -y @clawcall/mcpInstall ClawCall MCP in Gemini settings.json
~/.gemini/settings.json
{
"mcpServers": {
"clawcall": {
"command": "npx",
"args": [
"-y",
"@clawcall/mcp"
],
"env": {
"CLAWCALL_API_KEY": "<YOUR_CLAWCALL_API_KEY>"
}
}
}
}Configuration
CLAWCALL_API_KEYSecretOptional ClawCall API key (clawcall_sk_...). Auto-provisioned on the first call if unset; shadows the saved key file when present.
CLAWCALL_BASE_URLClawCall REST API base URL. Defaults to https://api.clawcall.dev.
| Name | Description | |
|---|---|---|
| CLAWCALL_API_KEY | Optional ClawCall API key (clawcall_sk_...). Auto-provisioned on the first call if unset; shadows the saved key file when present. | Secret |
| CLAWCALL_BASE_URL | ClawCall REST API base URL. Defaults to https://api.clawcall.dev. |
Automated checks
Ran Nov 13, 2026Full report →Passed our automated checks. That is not a security audit: read the code and give the server the narrowest access it needs. How checks work
- Source repository linked and reachableclawcall-dev/clawcall is public and reachablegithub.com/clawcall-dev/clawcall ↗Pass
- Package names this repository as its source@clawcall/mcp names clawcall-dev/clawcall as sourcenpmjs.com/package/@clawcall/mcp ↗Pass
- Build provenance (Sigstore attestation)No build attestation published for this releaseNot checked
- Open source license (OSI)MITspdx.org/licenses/MIT.html ↗Pass
- Commit or release in the last yearLast commit or release 2026-10-13github.com/clawcall-dev/clawcall ↗Pass
- OpenSSF ScorecardNot checked
- Package name not close to a popular packageNo popular package one edit away from @clawcall/mcpPass
- No critical advisory in the package or its dependencies99 resolved dependencies checked; no critical advisoryosv.dev ↗Pass
- No hidden instructions to the model0 tool descriptions and the README read; no override phrases, credential paths or hidden charactersPass
- Destructive tools are labelledNot checked
- Remote endpoint answers the MCP handshakeHTTPS endpoint answered 401 (sign-in required)Pass
README
View on GitHub@clawcall/mcp
[!IMPORTANT] Legacy local transport. New integrations should use ClawCall's hosted OAuth MCP connector at
https://api.clawcall.dev/mcp. This stdio package remains available for existing local/API-key installations, but it is not the connector submitted to the official xAI marketplace.
MCP server wrapping the ClawCall REST API for AI-powered US phone calls.
ClawCall places real outbound phone calls in the US (+1 only). This package is a stdio
MCP server that exposes the ClawCall REST contract as tools any MCP host (Claude Code,
Claude Desktop, Cursor, Windsurf) can call. It is a stateless client over the public API
at https://api.clawcall.dev — it imports nothing from the ClawCall server.
Install / register
Add the server to your MCP client config:
{ "mcpServers": { "clawcall": { "command": "npx", "args": ["-y", "@clawcall/mcp"], "env": { "CLAWCALL_API_KEY": "optional", "CLAWCALL_BASE_URL": "https://api.clawcall.dev" } } } }
Or, for Claude Code, the one-liner:
claude mcp add clawcall -- npx -y @clawcall/mcp
The env block is optional. The ClawCall server self-provisions an API key on the
first call: an anonymous place_call auto-issues a proto-key, which this server captures
and persists. You do not need to set CLAWCALL_API_KEY to get started.
Identity sharing
This MCP server reads and writes the same ~/.config/clawcall/key.json file
({ "api_key": "...", "user_phone_number": "..." }) that the ClawCall agent skill uses.
Running both the skill and this MCP server does not fork your identity — they share one
key, one trial quota, one call history, and one balance.
If CLAWCALL_API_KEY is set in the environment, it shadows the file (an operator /
CI override). The saved file is otherwise the source of truth, and the proto-key issued on
your first call is persisted there automatically (preserving any saved user_phone_number).
The workflow
A call runs for several minutes. The load-bearing loop is:
place_call— returns acall_idimmediately. Does not block.- Poll
get_callevery 3 seconds untillifecycle === "finalized". outcomeis the phone-network result (answered / no_answer / busy / …), NOT task success. An answered call can still fail the task.- Read the transcript with
get_call_transcriptand judge it against the task you set.
Choosing a tool
place_call_and_waitis the single-foreground-call shortcut: it places the call and internally polls until finalized or a hard cap (~180 s, ceiling 240 s). On the cap it degrades to the non-blocking contract — it returns{ still_running: true, call_id, … }and you continue withget_call. Use it when you want one call's result in one shot.- Campaigns — 3–4 parallel, interchangeable, info-gathering-only calls (quotes,
availability, inventory, policy comparison) — use
place_call+get_callfan-out. Never use the blocking tool for campaigns. Never parallelize calls that book / buy / cancel / commit.
Caveats
greetingdoes NOT set the agent's first spoken line. It is outbound-opener context only. The agent's actual opening line is a hardcoded recording disclosure on the server.- 402 errors are PERMANENT gating, not retryable.
trial_exhausted,plan_required, andreserved_number_requiredmean a human must act. Forward the returnedaction.url/sign_in_urlverbatim (they are personalized and embed your proto-key). Do not retry.
Trial limits are read from the live quota object in the error envelope — they are
never hardcoded in this package, so the displayed numbers always match the server.
Environment variables
| Variable | Default | Purpose |
|---|---|---|
CLAWCALL_BASE_URL |
https://api.clawcall.dev |
ClawCall REST API base URL. |
CLAWCALL_FRONTEND_URL |
https://clawcall.dev |
Frontend base used to build sign-in / account-link URLs. |
CLAWCALL_API_KEY |
(unset) | Optional operator override. Shadows the saved key file. |
CLAWCALL_WAIT_DEFAULT |
180 |
Default cap (seconds) for place_call_and_wait. |
CLAWCALL_WAIT_CEILING |
240 |
Maximum allowed cap (seconds) for place_call_and_wait. |
More
skill-companion/MCP-COMPANION.md— companion note for the ClawCall skill describing how the two share identity and workflow.../skill/SKILL.md— the ClawCall agent skill (the procedural layer; this MCP server is the connectivity layer over the same REST API).
ClawCall MCP questions
How do I add ClawCall MCP to Claude Code?
Run claude mcp add --env 'CLAWCALL_API_KEY=<YOUR_CLAWCALL_API_KEY>' clawcall -- npx -y @clawcall/mcp in your terminal. Add --scope user to make it available in every project.
How do I add ClawCall MCP to Cursor?
Open ~/.cursor/mcp.json and add the server under mcpServers, using the Cursor config shown in the install section above.
Do I need an API key to start using ClawCall?
No, the server can self-provision an API key. An anonymous call will auto-issue a proto-key which the server then captures and persists.
Can I make multiple calls at once?
Yes, you can use a fan-out approach with place_call and get_call for parallel info-gathering campaigns, but you should avoid parallelizing calls that involve booking, buying, or canceling.
How do I get the transcript of a call?
You can use the get_call_transcript tool to read the transcript after the call has been finalized.