[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M76Z4KFDFHNXYYJDA0WHW5XR":3,"skill-same:01M76Z4KFDFHNXYYJDA0WHW5XR":72},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":16,"categories":17,"keywords":18,"publishedAt":19,"linkOnly":20,"markdownContent":21,"markdownBody":22,"frontMatter":23,"refName":49,"commitSha":50,"skillDirectory":51,"forksCount":52,"repositoryHtmlUrl":53,"repositoryHomepageUrl":54,"ownerHtmlUrl":55,"ownerBlogUrl":6,"ownerDescription":56,"updatedAt":57,"install":58,"resources":71},"01M76Z4KFDFHNXYYJDA0WHW5XR","amber-voice-assistant",null,"Give your OpenClaw agent real phone capabilities. Setup uses a short wizard; after setup, run calls and phone tasks with one natural-language prompt.","Lord1Egypt\u002Fawesome-skill-forge","Lord1Egypt","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F99746359?v=4","01M7592YDNT0XK9WJ0T2MQVPDN","Mohamed Mounir Ali AL Keshita","01M75GRFNHC8VZNWEB8DKDE433",3,"MIT","https:\u002F\u002Fgithub.com\u002FLord1Egypt\u002Fawesome-skill-forge\u002Fblob\u002Fmaster\u002Fcommunity\u002Fclawhub\u002Fa\u002Famber-phone-capabilities\u002FSKILL.md",[],[],"2026-11-12 13:55:43",false,"---\nname: amber-voice-assistant\ntitle: \"Amber — Give Your Agent Real Phone Capabilities\"\ndescription: \"Give your OpenClaw agent real phone capabilities. Setup uses a short wizard; after setup, run calls and phone tasks with one natural-language prompt.\"\nhomepage: https:\u002F\u002Fgithub.com\u002Fbatthis\u002Famber-openclaw-voice-agent\nmetadata: {\"openclaw\":{\"emoji\":\"☎️\",\"requires\":{\"env\":[],\"optionalEnv\":[\"AMBER_ENABLE_OUTBOUND_CALLS\",\"OPENCLAW_GATEWAY_URL\",\"TWILIO_WEBHOOK_STRICT\",\"VOICE_PROVIDER\",\"VOICE_WEBHOOK_SECRET\",\"ASSISTANT_NAME\",\"OPERATOR_NAME\",\"AMBER_CRM_DB_PATH\",\"AGENT_MD_PATH\",\"DEFAULT_CALENDAR\"],\"anyBins\":[\"node\",\"ical-query\"]},\"install\":[{\"id\":\"runtime\",\"kind\":\"node\",\"cwd\":\"runtime\",\"label\":\"Install Amber runtime (cd runtime && npm ci && npm run build)\"}]}}\n---\n\n# Amber — Give Your Agent Real Phone Capabilities\n\n## Overview\n\nAmber gives any OpenClaw deployment **real phone capabilities for agents**. It ships with a **production-ready Twilio + OpenAI Realtime bridge** (`runtime\u002F`) for confirmed phone workflows: inbound answering, call screening, prepared outbound workflows, and confirmed scheduling over a real telephone number.\n\n**✨ New in v5.4.0:** Amber now ships as an **MCP plugin** with 9 tools — prepare confirmed calls by name, check call history, query CRM contacts, manage calendar, and control call screening. It works with Claude Desktop\u002FCowork and other MCP-capable clients or agent harnesses once configured. Includes Apple Contacts integration and a code-enforced call confirmation safeguard to prevent wrong-number dials.\n\n**✨ Also:** Interactive setup wizard (`npm run setup`) validates credentials in real-time and generates a working `.env` file — no manual configuration needed. Once setup is complete, Amber is prompt-based: ask your OpenClaw agent to prepare confirmed calls, answer\u002Fscreen callers, schedule confirmed appointments, or handle phone workflows in natural language.\n\n## See it in action\n\n![Setup Wizard Demo](demo\u002Fdemo.gif)\n\n**[▶️ Watch the interactive demo on asciinema.org](https:\u002F\u002Fasciinema.org\u002Fa\u002Fl1nOHktunybwAheQ)** (copyable text, adjustable speed)\n\n*The interactive wizard validates credentials, detects ngrok, and generates a complete `.env` file in minutes.*\n\n### What's included\n\n- **Runtime bridge** (`runtime\u002F`) — a complete Node.js server that connects Twilio phone calls to OpenAI Realtime with OpenClaw brain-in-the-loop\n- **Amber Skills** (`amber-skills\u002F`) — modular mid-call capabilities (CRM, calendar, log & forward message) with a spec for building your own\n- **Built-in CRM** — local SQLite contact database; Amber greets callers by name and references personal context naturally on every call, with operator review\u002Fcorrection responsibility\n- **Call log dashboard** (`dashboard\u002F`) — browse call history, transcripts, and captured messages; includes **manual Sync button** to pull new calls on demand\n- **Setup & validation scripts** — preflight checks, env templates, quickstart runner\n- **Architecture docs & troubleshooting** — call flow diagrams, common failure runbooks\n- **Safety guardrails** — outbound calls require code-enforced confirmation; payment escalation, consent boundaries, and explicit confirmation for calendar writes are documented\n\n## 🔌 Amber Skills — Extensible by Design\n\nAmber ships with a growing library of **Amber Skills** — modular capabilities that plug directly into live voice conversations. Each skill exposes a structured function that Amber can call mid-call, letting you compose powerful voice workflows without touching the bridge code.\n\n### 👤 CRM — Contact Memory *(v5.3.0)*\n\nAmber remembers every caller across calls and uses that memory to personalize every conversation.\n\n- **Runtime-managed** — lookup and logging happen automatically; Amber never has to \"remember\" to call CRM\n- **Personalized greeting** — known callers can be greeted by name; optional notes are used only when relevant to the call objective\n- **Two-pass enrichment** — auto-log captures the call immediately; an optional post-call extraction pass proposes caller details and notes for the local CRM\n- **Operator review expected** — review, correct, or delete CRM records periodically so bad transcript extraction or misleading caller input does not persist indefinitely\n- **Symmetric** — works identically for inbound and outbound calls\n- **Local SQLite CRM** — contact memory is stored at `~\u002F.config\u002Famber\u002Fcrm.sqlite`; CRM records are not cloud-hosted. Live call audio\u002Ftranscripts still pass through Twilio\u002FOpenAI as part of the phone bridge.\n- **Native dependency** — requires `better-sqlite3` (native build). macOS: `sudo xcodebuild -license accept` before `npm install`. Linux: `build-essential` + `python3`.\n\n### 📅 Calendar\n\nQuery the operator's calendar for availability or schedule a new event — all during a live call.\n\n- **Availability lookups** — free\u002Fbusy slots for today, tomorrow, this week, or any specific date\n- **Event creation** — create calendar events from a phone conversation only after the required details are collected and the caller explicitly confirms the slot\n- **Privacy by default** — callers are only told whether the operator is free or busy; event titles, names, and locations are never disclosed\n- Powered by `ical-query` — local-only, zero network latency\n\n### 📬 Log & Forward Message\n\nLet callers leave a message that is automatically saved and forwarded to the operator.\n\n- Captures the caller's message, name, and optional callback number\n- **Always saves to the call log first** (audit trail), then delivers via the operator's configured messaging channel\n- Confirmation-gated — Amber confirms with the caller before sending\n- Delivery destination is operator-configured — callers cannot redirect messages\n\n### Build Your Own Skills\n\nAmber's skill system is designed to grow. Each skill is a self-contained directory with a `SKILL.md` (metadata + function schema) and a `handler.js`. You can:\n\n- **Customize the included skills** to fit your own setup\n- **Build new skills** for your use case — CRM lookups, inventory checks, custom notifications, anything callable mid-call\n- **Share skills** with the OpenClaw community via [ClawHub](https:\u002F\u002Fclawhub.com)\n\nSee [`amber-skills\u002F`](amber-skills\u002F) for examples and the full specification to get started.\n\n> **Note:** Each skill's `handler.js` is reviewed against its declared permissions. When building or installing third-party skills, review the handler source as you would any Node.js module.\n\n### Call log dashboard\n\n```bash\ncd dashboard && node scripts\u002Fserve.js   # → http:\u002F\u002Flocalhost:8787\n```\n\n- **⬇ Sync button** (green) — immediately pulls new calls from `runtime\u002Flogs\u002F` and refreshes the dashboard. Use this right after a call ends rather than waiting for the background watcher.\n- **↻ Refresh button** (blue) — reloads existing data from disk without re-processing logs.\n- Background watcher (`node scripts\u002Fwatch.js`) auto-syncs every 30 seconds when running.\n\n## Why Amber\n\n- **Ship a voice assistant in minutes** — `npm install`, configure `.env`, `npm start`\n- Full inbound screening: greeting, message-taking, appointment booking with calendar integration\n- Outbound calls with structured call plans (reservations, inquiries, follow-ups), with confirmation gates and a runtime disable switch\n- **OpenClaw gateway lookup (least-privilege)** — voice agent consults your OpenClaw gateway only for call-critical needs (availability checks, confirmed scheduling, required factual lookups), not for unrelated tasks\n- VAD tuning + verbal fillers to keep conversations natural (no dead air during lookups)\n- Fully configurable: assistant name, operator info, org name, calendar, screening style — all via env vars\n- Operator safety guardrails for approvals\u002Fescalation\u002Fpayment handling\n\n## Personalization requirements\n\nBefore deploying, users must personalize:\n- assistant name\u002Fvoice and greeting text,\n- own Twilio number and account credentials,\n- own OpenAI project + webhook secret,\n- own OpenClaw gateway\u002Fsession endpoint,\n- own call safety policy (approval, escalation, payment handling).\n\nDo not reuse example values from another operator.\n\n## 5-minute quickstart\n\n### Option A: Interactive Setup Wizard (recommended) ✨\n\nThe easiest way to get started:\n\n1. `cd runtime`\n2. `npm run setup`\n3. Follow the interactive prompts — the wizard will:\n   - Validate your Twilio and OpenAI credentials in real-time\n   - Auto-detect and configure ngrok if available\n   - Generate a working `.env` file\n   - Optionally install dependencies and build the project\n4. Configure your Twilio webhook (wizard shows you the exact URL)\n5. Start the server: `npm start`\n6. Call your Twilio number — your voice assistant answers!\n\n**Benefits:**\n- Real-time credential validation (catch errors before you start)\n- No manual `.env` editing\n- Automatic ngrok detection and setup\n- Step-by-step guidance with helpful links\n\n### Option B: Manual setup\n\n1. `cd runtime && npm ci`\n2. Copy `..\u002Freferences\u002Fenv.example` to `runtime\u002F.env` and fill in your values.\n3. `npm run build && npm start`\n4. Point your Twilio voice webhook to `https:\u002F\u002F\u003Cyour-domain>\u002Ftwilio\u002Finbound`\n5. Call your Twilio number — your voice assistant answers!\n\n### Option C: Validation-only (existing setup)\n\n1. Copy `references\u002Fenv.example` to your own `.env` and replace placeholders.\n2. Export required variables (`TWILIO_ACCOUNT_SID`, `TWILIO_AUTH_TOKEN`, `TWILIO_CALLER_ID`, `OPENAI_API_KEY`, `OPENAI_PROJECT_ID`, `OPENAI_WEBHOOK_SECRET`, `PUBLIC_BASE_URL`).\n3. Run quick setup:\n   `scripts\u002Fsetup_quickstart.sh`\n4. If preflight passes, run one inbound and one outbound smoke test.\n5. Only then move to production usage.\n\n## Credential scope (recommended hardening)\n\nUse least-privilege credentials for every provider:\n\n- **Twilio:** use a dedicated subaccount for Amber and rotate auth tokens regularly.\n- **OpenAI:** use a dedicated project API key for this runtime only; avoid reusing keys from unrelated apps.\n- **OpenClaw Gateway token:** only set `OPENCLAW_GATEWAY_TOKEN` if you need brain-in-the-loop lookups; keep token scope minimal.\n- **Dependency integrity:** runtime dependencies are pinned by `runtime\u002Fpackage-lock.json`; review dependency changes before publishing updates.\n- **Secrets in logs:** never print full credentials in scripts, setup output, or call transcripts.\n- **Setup wizard validation scope:** credential checks call only official Twilio\u002FOpenAI API endpoints over HTTPS for auth verification; no arbitrary exfiltration endpoints are used.\n\nThese controls reduce blast radius if a host or config file is exposed.\n\n## Safe defaults\n\n- Outbound calling is enabled by default for the full phone-agent experience. Set `AMBER_ENABLE_OUTBOUND_CALLS=false` to disable the outbound call endpoint.\n- Require explicit approval before outbound calls. **Note on confirmation enforcement:** For MCP-initiated outbound calls (`make_call`), confirmation is enforced at the MCP server layer in code (the tool returns a preview and requires `confirmed=true` on a second call before dialing) — this is not LLM-only instruction. The LLM instruction layer provides an additional reminder, but the code gate is the primary enforcement mechanism.\n- If payment\u002Fdeposit is requested, stop and escalate to the human operator.\n- Keep greeting short and clear.\n- Use timeout + graceful fallback when `ask_openclaw` is slow\u002Funavailable.\n\n## Workflow\n\n1. **Confirm scope for V1**\n   - Include only stable behavior: call flow, bridge behavior, fallback behavior, and setup steps.\n   - Exclude machine-specific secrets and private paths.\n\n2. **Document architecture + limits**\n   - Read `references\u002Farchitecture.md`.\n   - Keep claims realistic (latency varies; memory lookups are best-effort).\n\n3. **Run release checklist**\n   - Read `references\u002Frelease-checklist.md`.\n   - Validate config placeholders, safety guardrails, and failure handling.\n\n4. **Smoke-check runtime assumptions**\n   - Run `scripts\u002Fvalidate_voice_env.sh` on the target host.\n   - Fix missing env\u002Fconfig before publishing.\n\n5. **Publish**\n   - Publish to ClawHub (example):  \n     `clawhub publish \u003Cskill-folder> --slug amber-voice-assistant --name \"Amber Voice Assistant\" --version 1.0.0 --tags latest --changelog \"Initial public release\"`\n   - Optional: run your local skill validator\u002Fpackager before publishing.\n\n6. **Ship updates**\n   - Publish new semver versions (`1.0.1`, `1.1.0`, `2.0.0`) with changelogs.\n   - Keep `latest` on the recommended version.\n\n## Troubleshooting (common)\n\n- **\"Missing env vars\"** → re-check `.env` values and re-run `scripts\u002Fvalidate_voice_env.sh`.\n- **\"Call connects but assistant is silent\"** → verify TTS model setting and provider auth.\n- **\"ask_openclaw timeout\"** → verify gateway URL\u002Ftoken and increase timeout conservatively.\n- **\"Webhook unreachable\"** → verify tunnel\u002Fdomain and Twilio webhook target.\n\n## Guardrails for public release\n\n- Never publish secrets, tokens, phone numbers, webhook URLs with credentials, or personal data.\n- Include explicit safety rules for outbound calls, payments, and escalation.\n- Mark V1 as beta if conversational quality\u002Flatency tuning is ongoing.\n\n## Install safety notes\n\n- Amber does **not** execute arbitrary install-time scripts from this repository.\n- Runtime install uses standard Node dependency installation in `runtime\u002F`, with dependencies pinned in `runtime\u002Fpackage-lock.json`.\n- CRM uses `better-sqlite3` (native module), which compiles locally on your machine.\n- Review `runtime\u002Fpackage.json` dependencies before deployment in regulated environments.\n\n## Resources\n\n- **Runtime bridge:** `runtime\u002F` (full source + README)\n- Architecture and behavior notes: `references\u002Farchitecture.md`\n- Release gate: `references\u002Frelease-checklist.md`\n- Env template: `references\u002Fenv.example`\n- Quick setup runner: `scripts\u002Fsetup_quickstart.sh`\n- Env\u002Fconfig validator: `scripts\u002Fvalidate_voice_env.sh`\n","# Amber — Give Your Agent Real Phone Capabilities\n\n## Overview\n\nAmber gives any OpenClaw deployment **real phone capabilities for agents**. It ships with a **production-ready Twilio + OpenAI Realtime bridge** (`runtime\u002F`) for confirmed phone workflows: inbound answering, call screening, prepared outbound workflows, and confirmed scheduling over a real telephone number.\n\n**✨ New in v5.4.0:** Amber now ships as an **MCP plugin** with 9 tools — prepare confirmed calls by name, check call history, query CRM contacts, manage calendar, and control call screening. It works with Claude Desktop\u002FCowork and other MCP-capable clients or agent harnesses once configured. Includes Apple Contacts integration and a code-enforced call confirmation safeguard to prevent wrong-number dials.\n\n**✨ Also:** Interactive setup wizard (`npm run setup`) validates credentials in real-time and generates a working `.env` file — no manual configuration needed. Once setup is complete, Amber is prompt-based: ask your OpenClaw agent to prepare confirmed calls, answer\u002Fscreen callers, schedule confirmed appointments, or handle phone workflows in natural language.\n\n## See it in action\n\n![Setup Wizard Demo](demo\u002Fdemo.gif)\n\n**[▶️ Watch the interactive demo on asciinema.org](https:\u002F\u002Fasciinema.org\u002Fa\u002Fl1nOHktunybwAheQ)** (copyable text, adjustable speed)\n\n*The interactive wizard validates credentials, detects ngrok, and generates a complete `.env` file in minutes.*\n\n### What's included\n\n- **Runtime bridge** (`runtime\u002F`) — a complete Node.js server that connects Twilio phone calls to OpenAI Realtime with OpenClaw brain-in-the-loop\n- **Amber Skills** (`amber-skills\u002F`) — modular mid-call capabilities (CRM, calendar, log & forward message) with a spec for building your own\n- **Built-in CRM** — local SQLite contact database; Amber greets callers by name and references personal context naturally on every call, with operator review\u002Fcorrection responsibility\n- **Call log dashboard** (`dashboard\u002F`) — browse call history, transcripts, and captured messages; includes **manual Sync button** to pull new calls on demand\n- **Setup & validation scripts** — preflight checks, env templates, quickstart runner\n- **Architecture docs & troubleshooting** — call flow diagrams, common failure runbooks\n- **Safety guardrails** — outbound calls require code-enforced confirmation; payment escalation, consent boundaries, and explicit confirmation for calendar writes are documented\n\n## 🔌 Amber Skills — Extensible by Design\n\nAmber ships with a growing library of **Amber Skills** — modular capabilities that plug directly into live voice conversations. Each skill exposes a structured function that Amber can call mid-call, letting you compose powerful voice workflows without touching the bridge code.\n\n### 👤 CRM — Contact Memory *(v5.3.0)*\n\nAmber remembers every caller across calls and uses that memory to personalize every conversation.\n\n- **Runtime-managed** — lookup and logging happen automatically; Amber never has to \"remember\" to call CRM\n- **Personalized greeting** — known callers can be greeted by name; optional notes are used only when relevant to the call objective\n- **Two-pass enrichment** — auto-log captures the call immediately; an optional post-call extraction pass proposes caller details and notes for the local CRM\n- **Operator review expected** — review, correct, or delete CRM records periodically so bad transcript extraction or misleading caller input does not persist indefinitely\n- **Symmetric** — works identically for inbound and outbound calls\n- **Local SQLite CRM** — contact memory is stored at `~\u002F.config\u002Famber\u002Fcrm.sqlite`; CRM records are not cloud-hosted. Live call audio\u002Ftranscripts still pass through Twilio\u002FOpenAI as part of the phone bridge.\n- **Native dependency** — requires `better-sqlite3` (native build). macOS: `sudo xcodebuild -license accept` before `npm install`. Linux: `build-essential` + `python3`.\n\n### 📅 Calendar\n\nQuery the operator's calendar for availability or schedule a new event — all during a live call.\n\n- **Availability lookups** — free\u002Fbusy slots for today, tomorrow, this week, or any specific date\n- **Event creation** — create calendar events from a phone conversation only after the required details are collected and the caller explicitly confirms the slot\n- **Privacy by default** — callers are only told whether the operator is free or busy; event titles, names, and locations are never disclosed\n- Powered by `ical-query` — local-only, zero network latency\n\n### 📬 Log & Forward Message\n\nLet callers leave a message that is automatically saved and forwarded to the operator.\n\n- Captures the caller's message, name, and optional callback number\n- **Always saves to the call log first** (audit trail), then delivers via the operator's configured messaging channel\n- Confirmation-gated — Amber confirms with the caller before sending\n- Delivery destination is operator-configured — callers cannot redirect messages\n\n### Build Your Own Skills\n\nAmber's skill system is designed to grow. Each skill is a self-contained directory with a `SKILL.md` (metadata + function schema) and a `handler.js`. You can:\n\n- **Customize the included skills** to fit your own setup\n- **Build new skills** for your use case — CRM lookups, inventory checks, custom notifications, anything callable mid-call\n- **Share skills** with the OpenClaw community via [ClawHub](https:\u002F\u002Fclawhub.com)\n\nSee [`amber-skills\u002F`](amber-skills\u002F) for examples and the full specification to get started.\n\n> **Note:** Each skill's `handler.js` is reviewed against its declared permissions. When building or installing third-party skills, review the handler source as you would any Node.js module.\n\n### Call log dashboard\n\n```bash\ncd dashboard && node scripts\u002Fserve.js   # → http:\u002F\u002Flocalhost:8787\n```\n\n- **⬇ Sync button** (green) — immediately pulls new calls from `runtime\u002Flogs\u002F` and refreshes the dashboard. Use this right after a call ends rather than waiting for the background watcher.\n- **↻ Refresh button** (blue) — reloads existing data from disk without re-processing logs.\n- Background watcher (`node scripts\u002Fwatch.js`) auto-syncs every 30 seconds when running.\n\n## Why Amber\n\n- **Ship a voice assistant in minutes** — `npm install`, configure `.env`, `npm start`\n- Full inbound screening: greeting, message-taking, appointment booking with calendar integration\n- Outbound calls with structured call plans (reservations, inquiries, follow-ups), with confirmation gates and a runtime disable switch\n- **OpenClaw gateway lookup (least-privilege)** — voice agent consults your OpenClaw gateway only for call-critical needs (availability checks, confirmed scheduling, required factual lookups), not for unrelated tasks\n- VAD tuning + verbal fillers to keep conversations natural (no dead air during lookups)\n- Fully configurable: assistant name, operator info, org name, calendar, screening style — all via env vars\n- Operator safety guardrails for approvals\u002Fescalation\u002Fpayment handling\n\n## Personalization requirements\n\nBefore deploying, users must personalize:\n- assistant name\u002Fvoice and greeting text,\n- own Twilio number and account credentials,\n- own OpenAI project + webhook secret,\n- own OpenClaw gateway\u002Fsession endpoint,\n- own call safety policy (approval, escalation, payment handling).\n\nDo not reuse example values from another operator.\n\n## 5-minute quickstart\n\n### Option A: Interactive Setup Wizard (recommended) ✨\n\nThe easiest way to get started:\n\n1. `cd runtime`\n2. `npm run setup`\n3. Follow the interactive prompts — the wizard will:\n   - Validate your Twilio and OpenAI credentials in real-time\n   - Auto-detect and configure ngrok if available\n   - Generate a working `.env` file\n   - Optionally install dependencies and build the project\n4. Configure your Twilio webhook (wizard shows you the exact URL)\n5. Start the server: `npm start`\n6. Call your Twilio number — your voice assistant answers!\n\n**Benefits:**\n- Real-time credential validation (catch errors before you start)\n- No manual `.env` editing\n- Automatic ngrok detection and setup\n- Step-by-step guidance with helpful links\n\n### Option B: Manual setup\n\n1. `cd runtime && npm ci`\n2. Copy `..\u002Freferences\u002Fenv.example` to `runtime\u002F.env` and fill in your values.\n3. `npm run build && npm start`\n4. Point your Twilio voice webhook to `https:\u002F\u002F\u003Cyour-domain>\u002Ftwilio\u002Finbound`\n5. Call your Twilio number — your voice assistant answers!\n\n### Option C: Validation-only (existing setup)\n\n1. Copy `references\u002Fenv.example` to your own `.env` and replace placeholders.\n2. Export required variables (`TWILIO_ACCOUNT_SID`, `TWILIO_AUTH_TOKEN`, `TWILIO_CALLER_ID`, `OPENAI_API_KEY`, `OPENAI_PROJECT_ID`, `OPENAI_WEBHOOK_SECRET`, `PUBLIC_BASE_URL`).\n3. Run quick setup:\n   `scripts\u002Fsetup_quickstart.sh`\n4. If preflight passes, run one inbound and one outbound smoke test.\n5. Only then move to production usage.\n\n## Credential scope (recommended hardening)\n\nUse least-privilege credentials for every provider:\n\n- **Twilio:** use a dedicated subaccount for Amber and rotate auth tokens regularly.\n- **OpenAI:** use a dedicated project API key for this runtime only; avoid reusing keys from unrelated apps.\n- **OpenClaw Gateway token:** only set `OPENCLAW_GATEWAY_TOKEN` if you need brain-in-the-loop lookups; keep token scope minimal.\n- **Dependency integrity:** runtime dependencies are pinned by `runtime\u002Fpackage-lock.json`; review dependency changes before publishing updates.\n- **Secrets in logs:** never print full credentials in scripts, setup output, or call transcripts.\n- **Setup wizard validation scope:** credential checks call only official Twilio\u002FOpenAI API endpoints over HTTPS for auth verification; no arbitrary exfiltration endpoints are used.\n\nThese controls reduce blast radius if a host or config file is exposed.\n\n## Safe defaults\n\n- Outbound calling is enabled by default for the full phone-agent experience. Set `AMBER_ENABLE_OUTBOUND_CALLS=false` to disable the outbound call endpoint.\n- Require explicit approval before outbound calls. **Note on confirmation enforcement:** For MCP-initiated outbound calls (`make_call`), confirmation is enforced at the MCP server layer in code (the tool returns a preview and requires `confirmed=true` on a second call before dialing) — this is not LLM-only instruction. The LLM instruction layer provides an additional reminder, but the code gate is the primary enforcement mechanism.\n- If payment\u002Fdeposit is requested, stop and escalate to the human operator.\n- Keep greeting short and clear.\n- Use timeout + graceful fallback when `ask_openclaw` is slow\u002Funavailable.\n\n## Workflow\n\n1. **Confirm scope for V1**\n   - Include only stable behavior: call flow, bridge behavior, fallback behavior, and setup steps.\n   - Exclude machine-specific secrets and private paths.\n\n2. **Document architecture + limits**\n   - Read `references\u002Farchitecture.md`.\n   - Keep claims realistic (latency varies; memory lookups are best-effort).\n\n3. **Run release checklist**\n   - Read `references\u002Frelease-checklist.md`.\n   - Validate config placeholders, safety guardrails, and failure handling.\n\n4. **Smoke-check runtime assumptions**\n   - Run `scripts\u002Fvalidate_voice_env.sh` on the target host.\n   - Fix missing env\u002Fconfig before publishing.\n\n5. **Publish**\n   - Publish to ClawHub (example):  \n     `clawhub publish \u003Cskill-folder> --slug amber-voice-assistant --name \"Amber Voice Assistant\" --version 1.0.0 --tags latest --changelog \"Initial public release\"`\n   - Optional: run your local skill validator\u002Fpackager before publishing.\n\n6. **Ship updates**\n   - Publish new semver versions (`1.0.1`, `1.1.0`, `2.0.0`) with changelogs.\n   - Keep `latest` on the recommended version.\n\n## Troubleshooting (common)\n\n- **\"Missing env vars\"** → re-check `.env` values and re-run `scripts\u002Fvalidate_voice_env.sh`.\n- **\"Call connects but assistant is silent\"** → verify TTS model setting and provider auth.\n- **\"ask_openclaw timeout\"** → verify gateway URL\u002Ftoken and increase timeout conservatively.\n- **\"Webhook unreachable\"** → verify tunnel\u002Fdomain and Twilio webhook target.\n\n## Guardrails for public release\n\n- Never publish secrets, tokens, phone numbers, webhook URLs with credentials, or personal data.\n- Include explicit safety rules for outbound calls, payments, and escalation.\n- Mark V1 as beta if conversational quality\u002Flatency tuning is ongoing.\n\n## Install safety notes\n\n- Amber does **not** execute arbitrary install-time scripts from this repository.\n- Runtime install uses standard Node dependency installation in `runtime\u002F`, with dependencies pinned in `runtime\u002Fpackage-lock.json`.\n- CRM uses `better-sqlite3` (native module), which compiles locally on your machine.\n- Review `runtime\u002Fpackage.json` dependencies before deployment in regulated environments.\n\n## Resources\n\n- **Runtime bridge:** `runtime\u002F` (full source + README)\n- Architecture and behavior notes: `references\u002Farchitecture.md`\n- Release gate: `references\u002Frelease-checklist.md`\n- Env template: `references\u002Fenv.example`\n- Quick setup runner: `scripts\u002Fsetup_quickstart.sh`\n- Env\u002Fconfig validator: `scripts\u002Fvalidate_voice_env.sh`\n",{"name":5,"title":24,"homepage":25,"metadata":26,"description":7},"Amber — Give Your Agent Real Phone Capabilities","https:\u002F\u002Fgithub.com\u002Fbatthis\u002Famber-openclaw-voice-agent",{"openclaw":27},{"emoji":28,"install":29,"requires":34},"☎️",[30],{"id":31,"cwd":31,"kind":32,"label":33},"runtime","node","Install Amber runtime (cd runtime && npm ci && npm run build)",{"env":35,"anyBins":36,"optionalEnv":38},[],[32,37],"ical-query",[39,40,41,42,43,44,45,46,47,48],"AMBER_ENABLE_OUTBOUND_CALLS","OPENCLAW_GATEWAY_URL","TWILIO_WEBHOOK_STRICT","VOICE_PROVIDER","VOICE_WEBHOOK_SECRET","ASSISTANT_NAME","OPERATOR_NAME","AMBER_CRM_DB_PATH","AGENT_MD_PATH","DEFAULT_CALENDAR","master","4aa004da461d820f4ea5499e58eb8d196cf99016","community\u002Fclawhub\u002Fa\u002Famber-phone-capabilities",0,"https:\u002F\u002Fgithub.com\u002FLord1Egypt\u002Fawesome-skill-forge","https:\u002F\u002Fskill-explorer-rho.vercel.app","https:\u002F\u002Fgithub.com\u002FLord1Egypt","Unbreakable. Unstoppable. Unmatched⚡.\r\n Speaker of machine tongues 🤖  | World-builder of decentralized dreams 🌐 | Defender of the digital realm🛡️.","2026-06-10 08:21:01",{"source":8,"skillName":5,"commands":59,"prompt":69,"zipAvailable":70},[60,63,66],{"tool":61,"command":62},"npx","npx skills add Lord1Egypt\u002Fawesome-skill-forge --skill amber-voice-assistant",{"tool":64,"command":65},"bunx","bunx skills add Lord1Egypt\u002Fawesome-skill-forge --skill amber-voice-assistant",{"tool":67,"command":68},"pnpm","pnpm dlx skills add Lord1Egypt\u002Fawesome-skill-forge --skill amber-voice-assistant","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002FLord1Egypt\u002Fawesome-skill-forge\u002Fblob\u002Fmaster\u002Fcommunity\u002Fclawhub\u002Fa\u002Famber-phone-capabilities\u002FSKILL.md",true,[],[73,81,88,95,102,109],{"skillNo":74,"name":75,"nameZh":6,"description":76,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":77,"categories":78,"keywords":79,"publishedAt":80},"01M785WKJB99G3STR1NF9NRTJW","youtube-content","YouTube transcripts to summaries, threads, blogs.","https:\u002F\u002Fgithub.com\u002FLord1Egypt\u002Fawesome-skill-forge\u002Fblob\u002Fmaster\u002Fskills\u002Fyoutube-content\u002FSKILL.md",[],[],"2026-11-15 14:19:17",{"skillNo":82,"name":83,"nameZh":6,"description":84,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":85,"categories":86,"keywords":87,"publishedAt":80},"01M785WEFHB0G2S9YA6EPR3H0E","xurl","X\u002FTwitter via xurl CLI: post, search, DM, media, v2 API.","https:\u002F\u002Fgithub.com\u002FLord1Egypt\u002Fawesome-skill-forge\u002Fblob\u002Fmaster\u002Fskills\u002Fxurl\u002FSKILL.md",[],[],{"skillNo":89,"name":90,"nameZh":6,"description":91,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":92,"categories":93,"keywords":94,"publishedAt":80},"01M785W98EDN4EFJXJPMQEHCRC","touchdesigner-mcp","Control a running TouchDesigner instance via twozero MCP — create operators, set parameters, wire connections, execute Python, build real-time visuals. 36 native tools.","https:\u002F\u002Fgithub.com\u002FLord1Egypt\u002Fawesome-skill-forge\u002Fblob\u002Fmaster\u002Fskills\u002Ftouchdesigner-mcp\u002FSKILL.md",[],[],{"skillNo":96,"name":97,"nameZh":6,"description":98,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":99,"categories":100,"keywords":101,"publishedAt":80},"01M785W5S9FP3QBH2V191E025P","test-driven-development","TDD: enforce RED-GREEN-REFACTOR, tests before code.","https:\u002F\u002Fgithub.com\u002FLord1Egypt\u002Fawesome-skill-forge\u002Fblob\u002Fmaster\u002Fskills\u002Ftest-driven-development\u002FSKILL.md",[],[],{"skillNo":103,"name":104,"nameZh":6,"description":105,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":106,"categories":107,"keywords":108,"publishedAt":80},"01M785W2AS08NSSH9YATCV6FJB","teams-meeting-pipeline","Operate the Teams meeting summary pipeline via Hermes CLI — summarize meetings, inspect pipeline status, replay jobs, manage Microsoft Graph subscriptions.","https:\u002F\u002Fgithub.com\u002FLord1Egypt\u002Fawesome-skill-forge\u002Fblob\u002Fmaster\u002Fskills\u002Fteams-meeting-pipeline\u002FSKILL.md",[],[],{"skillNo":110,"name":111,"nameZh":6,"description":112,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":113,"categories":114,"keywords":115,"publishedAt":80},"01M785VXKW4T34BTP1SCHW1R7F","systematic-debugging","4-phase root cause debugging: understand bugs before fixing.","https:\u002F\u002Fgithub.com\u002FLord1Egypt\u002Fawesome-skill-forge\u002Fblob\u002Fmaster\u002Fskills\u002Fsystematic-debugging\u002FSKILL.md",[],[]]