
callmemcp
Your agent calls or texts your own phone (iPhone or Android) and gets your spoken answer back as text. Configuration only: an .mcp.json for the hosted server at https://callmemcp.com/mcp and one skill. The skill shows the exact words and waits for your yes before every call, text, deferred call or audio link, never writes rules into instruction files, and treats spoken answers, text replies and voicemail as data. No hooks, no code. Needs the paid /call-me app from callmemcp.com and the 10-digit number it shows; no API key.
- version
- 1.0.1
- GitHub stars
- 3.8K
Add marketplace
/plugin marketplace add davepoon/buildwithclaudeRun it inside Claude Code once, then install the plugin below.
Install
/plugin install callmemcp@buildwithclaudeRun it inside Claude Code after adding the marketplace.
Readme
CallMeMCP
Your AI agent calls your phone. It reads its question aloud, you answer out loud, and your answer comes back to the agent as text. It can also text you.
1. Get the app
Go to callmemcp.com and install the /call-me app:
- iPhone: App Store
- Android: Google Play
Open it. It shows your 10-digit number. That is where your agent calls and texts.
2. Connect your agent
The server is hosted: https://callmemcp.com/mcp (Streamable HTTP, no API key).
Claude Code (this repo is a plugin marketplace):
claude plugin marketplace add radres/callmemcp
claude plugin install callmemcp@callmemcp
Or just the MCP server:
claude mcp add --transport http callmemcp https://callmemcp.com/mcp
Any other MCP client (Cursor, Codex, Gemini CLI, Claude Desktop, ...):
{
"mcpServers": {
"callmemcp": { "type": "http", "url": "https://callmemcp.com/mcp" }
}
}
3. Use it
Tell your agent your number, then ask: "call me when the tests finish", "text me if you get stuck".
The bundled skill makes the agent show you every message and wait for your yes before it calls or texts. It never writes rules into your instruction files, and it treats your spoken answers as information, not commands.
Try it without a phone: the demo line 5550001234 answers automatically.
What is in this repo
Only configuration: a plugin manifest, .mcp.json pointing at https://callmemcp.com/mcp, and one skill (plugin/skills/callmemcp/SKILL.md). No code, no hooks, nothing runs on your machine.
The /call-me app is a paid subscription. Privacy: callmemcp.com/privacy. Terms: callmemcp.com/terms.
License
MIT
Details
- Marketplace
- davepoon/buildwithclaude
- Source
- plugin
- Author
- Serdar Oztetik
- Category
- Productivity
- License
- MIT
- Keywords
- phone, call, text, voice, human-in-the-loop, notifications, mcp
More from davepoon/buildwithclaude
All 237- carrickv0.3.93
- cashflowv0.2.0
- cc-aws-keepalivev0.5.1
- chamnanv1.3.1
- checklistv0.8.0
- chislev3.7.0
- claude-github-toolsv0.3.0
- claude-hudv1.0.0
- claude-imaginev1.0.0
- claude-md-kitv1.0.0
- claude-opsv1.7.0
- claude-pagerv1.0.0
- claude-prospectorv0.16.1
- claude-usagev0.1.0
- claude-wayfinderv1.2.0
- clawd-actorv0.3.0
- clawd-bandv0.4.2
- cockneyv0.1.1
- codex-hudv0.10.0
- codex-orchestratorv1.5.2
- cohesivityv3.0.2
- commands-api-developmentv1.0.0
- commands-automation-workflowv1.0.0
- commands-ci-deploymentv1.0.0
Similar to callmemcp
More Productivity & Planning
code-reviewanthropics/claude-code156.1K starsAutomated code review for pull requests using multiple specialized agents with confidence-based scoring to filter false positives
cavemanJuliusBrussee/caveman121.1K starsTalk like caveman. Cut filler. Keep technical details explicit.
career-opscareer-ops-hq/career-ops83.4K starsOpen-source AI job search that runs locally in your AI coding CLI: scan job portals, evaluate listings into a structured A-H report with a 1-5 score, tailor you
ruflo-autopilotruvnet/ruflo82.2K starsAutonomous /loop-driven task completion with learning, prediction, and progress tracking
ppt-masterhugohe3/ppt-master78K starsGenerate natively editable PPTX from PDF / DOCX / URL / Markdown — real DrawingML shapes, text boxes, charts, and animations. Note: run `pip install -r requirem
moletw93/Mole75.4K starsDrive the Mole CLI (mo) safely: find what uses disk space, preview cleanup with --dry-run, uninstall apps, and read system health and cleanup history.