Awesome Phone Call Agents is an early community hub for developers and workflow builders creating reusable phone-call workflows for AI agents. CALL-E SDKs, provider APIs, authentication, call execution, and provider-side controls belong upstream with CALL-E itself. This repository focuses on the community artifacts around those primitives: Agent Skills, workflow plugins, user-facing apps, examples, templates, and safety patterns.

The community roadmap is a direction guide, not a fixed release plan. Small examples, platform notes, workflow sketches, templates, and focused demos are all useful.

• Why this repository exists
• Quick install and start
• Community contributions
• What this repository provides
• Reference skills
• CLI reference
• Templates
• Resource list
• Apps
• Plugins
• Repository layout
• Safety and legal guide
• Contributing
• Developer docs
• Community
• License

AI agents increasingly need to turn phone calls into reusable workflows: reminders, follow-ups, appointment coordination, provider-specific call adapters, scheduler integrations, runnable demo apps, safety checks, and reference apps that other agents can install or adapt.

This repository exists to collect those phone-call capabilities and scenarios as portable Agent Skills, apps, adapters, scheduler recipes, and safety patterns. Each entry should help an agent package, schedule, execute, or safely operate a real phone-call workflow.

The scope is intentionally focused on AI-agent phone-call workflows, not generic voice-agent products, telephony vendor directories, or call-center software lists.

This repository focuses on three principles:

1. Portability: skills, apps, and adapters should be useful across agent hosts when possible.
2. Provider separation: the phone-call provider should place or create calls; the host scheduler should handle recurrence.
3. Safety by default: phone numbers, consent, credentials, and medical, legal, financial, or emergency boundaries must be handled explicitly.

Start from the resource list when you want a ready-to-use phone-call workflow, from apps when you want to study or run an integration pattern, from plugins when you want to connect workflow platforms, or from the templates when you want to contribute a new skill, app, plugin, adapter, scheduler recipe, or safety pattern.

The first reference skill is call-reminder, a daily reminder workflow that shows how to package one phone-call scenario as an installable skill with scheduling and safety boundaries. The google-form-callback skill shows how to process authorized Google Form responses into reviewed one-off callback calls with optional scheduled polling.

For most users, the simplest path is to ask an Agent Skills-compatible client to install the specific skill you want. To try the official daily reminder workflow, install call-reminder:

Install the portable call-reminder skill for this agent.

npx -y skills add CALLE-AI/awesome-phone-call-agents --skill call-reminder -g

After installation:
1. Reload or restart the agent if the client requires it.
2. Use the skill only for explicit phone-call reminder requests.
3. Follow the skill's scheduler, consent, cancellation, and credential-safety rules.

Manual copy is also supported when a host does not support the CLI:

skills/call-reminder/

Use a fictional reserved number in demos and a real E.164 number only for your own authorized workflow.

Set up a daily phone-call reminder at 09:00 America/New_York.
Use CALL-E when available.
My phone number is +15550101234.
The call should remind me to take my medicine according to my doctor instructions or the medication label.

The agent should create or update a visible scheduler job only when the current client can persist, run, and cancel it safely. The scheduled job then executes the runtime prompt and attempts exactly one one-off CALL-E call.

auth status -> call plan -> call run -> call status

If the client cannot safely create the schedule, the skill must return status: not created with the exact blocker and a runtime prompt or setup instructions.

call-reminder is a scheduler wrapper skill, not a new CALL-E backend reminder API, daemon, or provider-side recurring schedule.

• daily or recurring CALL-E phone reminders
• scheduled CALL-E calls where the scheduler belongs to the client, host, cron, or automation system
• "call me at a time" and "remind me by phone" workflows
• runtime prompt generation for scheduled one-off calls
• choosing the safest available scheduler adapter for the current client

• create CALL-E backend reminder APIs such as create_call_reminder, list_call_reminders, update_call_reminder, or cancel_call_reminder
• make provider-side recurrence mandatory
• install calle globally without explicit user approval
• guess phone numbers, country codes, timezones, languages, or regions
• create recurring calls to third-party numbers unless the user explicitly states that the recipient consented
• place a setup-time test call unless the user explicitly asks for one

CALL-E CLI parameters and command flags are documented in cli-reference.md.

The project-level validation script applies to the whole repository. The current Node.js helper scripts belong to the official call-reminder skill and show the expected shape for future skill-specific utilities.

Use this Agent Skills folder pattern:

skill-name/
├── SKILL.md
├── references/
├── scripts/
└── assets/

Use apps/ for runnable tools and demo apps:

apps/
├── python/
│   └── app-name/
└── typescript/
    └── app-name/

Every app that can place a call or create a recurring job must document setup, side effects, cancellation, credential handling, and dry-run or preview behavior.

Use plugins/ for no-code and low-code workflow-platform plugins:

plugins/
└── plugin-name/
    ├── README.md
    ├── manifest-or-config-file
    └── examples/

Every plugin should document supported triggers or actions, required inputs, side effects, credential handling, dry-run or preview behavior, and cancellation or rollback behavior when it can create calls or recurring jobs.

- [Project Name](https://example.com) - One sentence explaining why this is useful for AI-agent phone-call workflows.

Keep descriptions short, specific, factual, and directly tied to packaging, scheduling, executing, or safely operating AI-agent phone-call tasks.

This project is an awesome list for AI-agent phone-call workflows. Add resources only when they directly help agents package, schedule, execute, or safely operate phone-call tasks.

• call-reminder - Scheduler wrapper skill for recurring CALL-E phone-call reminders.
• google-form-callback - Google Form response workflow for safe one-off callback calls with dry-runs, scheduling plans, and Sheets writeback. See the workflow guide.

Runnable demo apps live under apps/. They are not a CALL-E SDK and do not define a supported application API.

The default e2e tests use a local fake broker/OAuth/MCP server or dry-run paths, so they do not require real CALL-E credentials or browser login. Live verification is opt-in in each app README.

No-code and low-code workflow plugins live under plugins/. They are for workflow-platform nodes, actions, connectors, and recipes that help operators connect business events to phone-call agent workflows without writing a full app.

Plugins should be explicit about inputs, outbound call side effects, credential handling, preview or dry-run behavior, and how a workflow builder can disable or roll back the integration.

• CALL-E CLI bootstrap - Resolver order for repository-local, global, and pinned npx CALL-E CLI routes.
• Client adapter matrix - Multi-client adapter guidance for CALL-E scheduled reminders.
• Runtime prompt template - Self-contained prompt used by scheduled jobs.
• Runtime prompt behavior checks - Behavior checks for scheduler selection, timezone handling, region handling, and provider boundaries.

• Safety reference - Consent, E.164 phone-number handling, credential boundaries, cancellation, duplicate-job prevention, and medical reminder boundaries.
• Design principles - Repository-wide architecture principles for safe phone-call workflows.

awesome-phone-call-agents/
├── README.md
├── AGENTS.md
├── CONTRIBUTING.md
├── SECURITY.md
├── apps/
│   ├── README.md
│   ├── python/
│   │   ├── batch-runner/
│   │   ├── broker-login-client/
│   │   └── oauth-login-client/
│   ├── shared/
│   │   └── fake-mcp-broker-server.mjs
│   └── typescript/
│       ├── broker-login-client/
│       ├── broker-login-client-standalone/
│       └── oauth-login-client/
├── docs/
│   ├── google-form-callback/
│   │   ├── README.md
│   │   └── assets/
│   ├── design-principles.md
│   ├── codex-implementation-plan.md
│   └── roadmap.md
├── forms/
│   └── callback-request/
├── plugins/
│   └── README.md
├── scripts/
│   └── validate_repository.py
└── skills/
    ├── call-reminder/
    │   ├── SKILL.md
    │   ├── references/
    │   └── scripts/
    └── google-form-callback/
        ├── SKILL.md
        ├── references/
        ├── scripts/
        └── template.md

Phone calls are real-world side effects. Preserve these rules across the whole project: skills, apps, provider adapters, scheduler recipes, automation patterns, reference implementations, and documentation.

• Require explicit user intent before setup or execution.
• Collect the required destination, timing, task, and consent details before creating a call workflow.
• Do not infer critical call details from phone number, locale, IP address, UTC offset, language, or country code.
• Mask phone numbers in user-facing summaries. Documentation examples should use reserved fictional numbers such as +15550101234.
• Do not call any number except the configured E.164 phone number.
• Do not modify user-provided call goals or messages except for safety-preserving formatting.
• Do not create duplicate scheduled jobs or hidden recurring schedules.
• For scheduled workflows, the visible scheduler job approval is the confirmation for future executions; runtime prompts must not require another per-run chat approval unless the job was explicitly configured as preview-only.
• Verify required auth before creating scheduled jobs, then re-check auth at runtime before placing any call.
• Do not create third-party recurring calls unless the user explicitly states recipient consent.
• Do not expose API keys, OAuth tokens, access tokens, refresh tokens, session cookies, auth callback URLs, confirmation tokens, or provider credentials.
• If auth is missing, the CLI is unavailable, the scheduler cannot access credentials, or required fields are ambiguous, skip the call or stop setup instead of guessing.
• Treat medical, legal, financial, and emergency reminders as logistics only.
• Every successful setup summary must include cancellation or update instructions when the selected adapter supports them.

This repository does not provide legal advice. Contributors and users are responsible for complying with laws, platform rules, consent requirements, call-recording rules, telemarketing restrictions, emergency-service boundaries, and provider terms that apply to their workflow and jurisdiction.

Do not submit resources that hide calls from users, hide recurring jobs, bypass consent, expose credentials, or encourage contacting people who did not authorize the workflow.

See CONTRIBUTING.md for the full contribution guide.

1. Choose a scoped contribution: skill, app, provider adapter, scheduler recipe, automation pattern, safety pattern, or reference implementation.
2. Confirm it directly helps AI agents package phone-call workflows.
3. Use the templates above for skill folders, app directories, adapter records, or README entries.
4. Add setup, usage, side-effect, and cancellation notes.
5. Use fictional or masked phone numbers in samples.
6. Keep repository-facing content in English.
7. Run validation before opening a pull request.

python3 scripts/validate_repository.py

High-quality additions should include a short description, compatibility notes, safety notes for real-world side effects, setup or install instructions, tests, cancellation or rollback behavior for recurring workflows, and no secrets or personal data.

Out of scope: generic telephony vendor directories, marketing-only pages, call-center software lists without an AI-agent workflow, tools that require unsafe credential handling, and resources that hide phone calls, recurring jobs, or external side effects from the user.

• Discord: https://discord.gg/6AbXUzUV8w

MIT. See LICENSE.