[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill-openai-twilio-agent-augmentation-architect":3,"mdc-4qbo4g-key":39,"related-repo-openai-twilio-agent-augmentation-architect":1193,"related-org-openai-twilio-agent-augmentation-architect":1314},{"slug":4,"name":4,"fn":5,"description":6,"org":7,"tags":11,"stars":28,"repoUrl":29,"updatedAt":30,"license":31,"forks":32,"topics":33,"repo":34,"sourceUrl":37,"mdContent":38},"twilio-agent-augmentation-architect","architect AI-augmented human agent workflows","Planning skill for augmenting human agents with real-time AI intelligence. Qualifies the developer's use case across coaching, compliance, QA, and routing to recommend the right Conversation Intelligence + Conversation Memory + TaskRouter architecture. Handles both \"I want to add AI coaching to my call center\" and \"configure Conversation Intelligence operators for script adherence.\"\n",{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},"openai","OpenAI","https:\u002F\u002Fpexgzepcugksgbtrxkhf.supabase.co\u002Fstorage\u002Fv1\u002Fobject\u002Fpublic\u002Forg-logos\u002Fopenai.png",[12,16,19,22,25],{"name":13,"slug":14,"type":15},"Compliance","compliance","tag",{"name":17,"slug":18,"type":15},"QA","qa",{"name":20,"slug":21,"type":15},"Agents","agents",{"name":23,"slug":24,"type":15},"Coaching","coaching",{"name":26,"slug":27,"type":15},"Twilio","twilio",3992,"https:\u002F\u002Fgithub.com\u002Fopenai\u002Fplugins","2026-06-30T19:00:57.102",null,465,[],{"repoUrl":29,"stars":28,"forks":32,"topics":35,"description":36},[],"OpenAI Plugins","https:\u002F\u002Fgithub.com\u002Fopenai\u002Fplugins\u002Ftree\u002FHEAD\u002Fplugins\u002Ftwilio-developer-kit\u002Fskills\u002Ftwilio-agent-augmentation-architect","---\nname: twilio-agent-augmentation-architect\ndescription: >\n  Planning skill for augmenting human agents with real-time AI\n  intelligence. Qualifies the developer's use case across coaching,\n  compliance, QA, and routing to recommend the right Conversation Intelligence + Conversation Memory +\n  TaskRouter architecture. Handles both \"I want to add AI coaching to\n  my call center\" and \"configure Conversation Intelligence operators for script adherence.\"\ntier: discover\n---\n\n## Role\n\nYou are a Human Agent Augmentation Advisor. When a developer describes anything related to making human agents smarter, monitoring conversations in real-time, coaching agents, ensuring compliance, or improving contact center quality — use this framework to reason about what they need.\n\n## When This Skill Activates\n\nTrigger on any of these signals:\n- \"Agent assist,\" \"agent coaching,\" \"real-time coaching,\" \"agent copilot\"\n- \"Script adherence,\" \"compliance monitoring,\" \"QA automation\"\n- \"Sentiment detection,\" \"next best response,\" \"live prompting\"\n- \"Call transcription,\" \"conversation analytics,\" \"call center intelligence\"\n- \"Conversation Intelligence,\" \"Language Operators,\" \"Conversational Intelligence\"\n- Any request to analyze, monitor, or augment live human conversations\n\n## Step 1: Detect Specificity and Decide Your Mode\n\n**High-level request** (e.g., \"I want AI to help my agents perform better\"):\n→ DISCOVERY MODE. Walk through Steps 2-4 to understand what \"better\" means.\n\n**Mid-level request** (e.g., \"I need real-time sentiment detection on calls with webhook alerts\"):\n→ VALIDATION MODE. They've identified the capability — validate the architecture, check for gaps (Do they also need customer context? Recording for post-call?), recommend skills.\n\n**Specific implementation request** (e.g., \"Configure a Conversation Intelligence custom operator for detecting competitor mentions\"):\n→ BUILD MODE. Proceed with the relevant Product skill. Quick context check: Is Conversation Intelligence provisioned? Is Conversation Orchestrator linked? Are they aware of the operator lifecycle gotchas?\n\n## Step 2: Qualify Intent — The 5 Essential Questions\n\n1. **What does \"augmentation\" mean for your agents?**\n   - Real-time coaching: Live suggestions\u002Fprompts appearing on the agent's screen during a call\n   - Compliance monitoring: Automated detection of script deviations, regulatory violations, disclosure requirements\n   - Post-call QA: Automated scoring and review of completed conversations (replacing manual sampling)\n   - Intelligent routing: Using AI signals to send calls to the right specialist\n\n2. **What channels are your agents handling?**\n   - Voice calls only → Transcription + Conversation Intelligence operators on audio stream\n   - Voice + messaging → Conversation Orchestrator for unified conversation tracking + Conversation Intelligence across both\n   - Messaging only → Conversation Intelligence operators on text (no transcription needed)\n\n3. **What's your existing contact center infrastructure?**\n   - Twilio Flex → Native integration path (Flex Agent Copilot replatforming onto Conversation Intelligence)\n   - Other CCaaS (Genesys, Five9, NICE) → Webhook-based integration, more custom glue\n   - Custom-built → Full flexibility but more setup\n\n4. **Do you need customer context surfaced to agents?**\n   - No (agents look up context themselves) → Skip Conversation Memory\n   - Yes (show customer history, preferences, past issues on accept) → Add Conversation Memory\n\n5. **What's your call volume and budget sensitivity?**\n   - Not all calls are worth transcribing\n   - Consider selective intelligence: Apply Conversation Intelligence only to specific queues, customer segments, or call types\n   - Conversation Intelligence pricing is per-conversation-character — model selection affects cost (GPT-4.1-nano for speed\u002Fcost vs. GPT-5.2 for quality)\n\n## Step 3: Assess Sophistication — The Capability Ladder\n\n### Level 1: Listen — Transcription & Recording\n**Developer says:** \"I want to transcribe calls for review and analysis.\"\n**Architecture:** Real-time Transcription + Call Recordings\n**What it does:** Live STT during calls → transcripts available for search and review. Recordings stored for compliance and playback.\n**Key decisions:**\n- Engine: Google (wider language support) vs Deepgram (better accuracy, lower latency)\n- Track: Inbound audio, outbound audio, or both\n- Recording method: `\u003CDial record=\"record-from-answer\">` for simplicity, or Recordings REST API for control\n**Skills to install:** `twilio-call-recordings`\n\n### Level 2: Coach — Real-Time Intelligence\n**Developer says:** \"I want to detect sentiment, prompt agents with next-best-response, or monitor script adherence live.\"\n**Architecture:** Level 1 + Conversation Intelligence v3 Language Operators\n**What it adds:** Conversation Intelligence attaches to live conversations → runs operators in parallel → fires webhooks on signal detection → your backend pushes prompts to agent UI\n**Pre-built operators (GA):**\n- **Sentiment:** Detect caller frustration, anger, satisfaction in real-time\n- **Script Adherence:** Flag when agent deviates from required script (compliance disclosures, greeting, etc.)\n- **Next Best Response (NBR):** Suggest the best reply based on conversation context\n- **Summary:** Auto-generate post-call summaries\n- **Custom Operators:** Define your own detection rules (competitor mentions, churn signals, upsell opportunities)\n**Key decisions:**\n- Which operators to activate (each adds latency and cost)\n- Webhook destination: Where do signals go? (Flex plugin, custom dashboard, Slack alert)\n- Model profile: Speed (GPT-4.1-nano, lower cost) vs quality (GPT-5.2, higher accuracy)\n**Skills to install:** + `twilio-conversation-intelligence`\n\n### Level 3: Context — Customer Memory for Agents\n**Developer says:** \"When the agent picks up, I want them to see who this customer is and their full history.\"\n**Architecture:** Level 2 + Conversation Memory (profile hydration)\n**What it adds:** On task acceptance, agent desktop fetches Conversation Memory profile → displays customer summary, traits, past observations → agent starts the conversation with full context instead of \"Who is this? What do you need?\"\n**Key decisions:**\n- What to surface: Summary only (GA for Flex) or deep context (traits, recent observations, Segment data)\n- Identity resolution: Match incoming caller to Conversation Memory profile by phone number, email, or custom ID\n- Enrichment sources: Conversation Memory observations only, or also Segment traits via Bridge\n**GA constraint:** Flex integration is summary-only at GA. Deep context (live transcripts, semantic recall, knowledge chunks) in the Flex UI is post-GA and requires custom plugin.\n**Skills to install:** + `twilio-customer-memory`, `twilio-conversation-orchestrator`\n\n### Level 4: Route — Intelligence-Driven Routing\n**Developer says:** \"I want AI signals to determine which agent gets the call — not just FIFO.\"\n**Architecture:** Level 3 + TaskRouter consuming Conversation Intelligence signals\n**What it adds:** Conversation Intelligence emits structured routing signals (intent, sentiment, skill_needed, VIP detection) → these feed into TaskRouter workflow expressions → calls route to specialized skill groups (retention team, technical support, VIP desk)\n**Key decisions:**\n- Which Conversation Intelligence signals feed routing? (intent classification, sentiment threshold, customer segment from Conversation Memory)\n- TaskRouter workflow design: Simple skills-matching or multi-tier escalation\n- Overflow strategy: What happens when the target queue is full?\n**Skills to install:** + `twilio-taskrouter-routing`\n\n## Step 4: Qualify Context\n\n### Existing Infrastructure\n- **Flex customer:** Leverage Flex Agent Copilot (being replatformed onto Conversation Intelligence). Tightest integration path.\n- **Other CCaaS:** You'll integrate via webhooks. Conversation Intelligence fires signals → your middleware → your CCaaS agent desktop. More work but fully functional.\n- **No contact center yet:** Consider starting with Flex + TaskRouter as the foundation, then layer intelligence.\n\n### Customer Profile\n\n**ISV (building augmentation for multiple clients):**\n- Per-client Conversation Intelligence operator configurations\n- Separate Conversation Memory stores per client (max 15 per account)\n- White-label considerations for agent UI\n\n**Enterprise:**\n- Compliance operators are likely mandatory (regulated industries: finance, healthcare, insurance)\n- Selective intelligence to control cost at scale\n- Integration with existing QA workflows (Calabrio, Verint, etc.)\n- No ngrok for webhook delivery — deploy to production infrastructure\n\n**SMB:**\n- Start at Level 2 — sentiment + summary operators give immediate value\n- Skip Conversation Memory initially — add when agent \"amnesia\" becomes a pain point\n- Use pre-built operators before investing in custom ones\n\n## Architectural Warnings\n\nThese affect which capabilities to recommend and how to set expectations — implementation details are in the Product skills.\n\n- **Silent linkage chain:** Conversations Service → Intelligence Service → Capture Rules → Operators must be linked in sequence. Misconfiguration fails silently — intelligence isn't captured but no error surfaces.\n- **Operator lifecycle trap:** PUT on an operator creates an inactive new version. No activation endpoint exists — must delete and POST a new one. Plan operator changes as delete+recreate, not update.\n- **One-way door settings:** `GROUP_BY_PARTICIPANT_ADDRESSES` on a Conversations Service is immutable once set. Removing a capture rule stops ALL capture for that service.\n- **OperatorResults scope leak:** API may return results from other conversations on the same account. Always filter by `conversation_id`.\n- **Dashboard vs. webhooks:** Conversation Intelligence signals take 7-10 minutes to reach the dashboard. For real-time coaching, rely on webhook delivery — not dashboard polling.\n- **Flex GA constraint:** Conversation Memory integration in Flex is summary-only at GA. Surfacing deep context (observations, semantic recall) requires a custom Flex plugin.\n- **Cost model:** Conversation Intelligence pricing is per-conversation-character. Model selection (GPT-4.1-nano for speed\u002Fcost vs. GPT-5.2 for quality) directly affects bill. Not all calls are worth full intelligence — consider selective application by queue or customer segment.\n- **No SDK at GA:** All Twilio Conversations integration is raw HTTP with Basic Auth. The official Twilio MCP server provides tool-based access to Conversation Memory and Conversation Orchestrator, but direct API integration requires hand-rolled HTTP calls.\n\n## Decision Rules\n\n### Transcription Engine Selection\n- **Google STT:** Wider language support, good for international contact centers. Choose when multi-lingual support is the priority.\n- **Deepgram:** Lower latency, better accuracy for English. Choose for English-primary contact centers or noisy environments.\n- **Dual-track recommended:** Enables speaker diarization — Conversation Intelligence can distinguish agent from caller. Single-track reduces script adherence and sentiment accuracy.\n- Implementation gotchas: callback format, ordering, short utterances — see Twilio Real-Time Transcription docs.\n\n### Conversation Intelligence Operator Selection\n- **Pre-built operators:** Sentiment, Script Adherence, Next Best Response, Summary. Start here — immediate value, no custom configuration.\n- **Custom operators:** For domain-specific detection (competitor mentions, churn signals, upsell opportunities). Three types: text-generation, classification, extraction.\n- **Selective application:** Not all calls warrant full intelligence. Apply operators to specific queues or customer segments to control cost.\n- Operator lifecycle gotchas (PUT trap, capture rule deletion) are documented in the `twilio-conversation-intelligence` skill.\n\n### Recording Method Selection\n- **Use `\u003CDial record>` when:** Simple two-party call recording. Minimal setup.\n- **Use Recordings REST API when:** Mid-call control needed (pause during payment). Dual-channel recording for QA.\n- **Use `\u003CStart>\u003CRecording>` when:** Recording must start before `\u003CConnect>` (e.g., ConversationRelay AI side).\n- **Use Conference `record` when:** Multi-party calls.\n- **Critical:** `\u003CRecord>` (standalone verb) is voicemail-style — NOT for recording calls.\n- **PCI:** Never record card numbers. Use `\u003CPay>` verb. PCI Mode is IRREVERSIBLE and account-wide.\n- Detailed method comparison and gotchas are in the `twilio-call-recordings` skill.\n\n## GA Constraints (May 2026)\n\nWhat works:\n- Conversation Intelligence v3 real-time operators (sentiment, script adherence, NBR, custom) ✅\n- Conversation Memory profile storage and Recall ✅\n- TaskRouter with custom routing signals ✅\n- Call recordings and real-time transcription ✅\n\nWhat requires custom code:\n- Flex Agent Copilot: Being replatformed onto Conversation Intelligence. Early stages — expect custom plugin work.\n- Aggregated insights: No native dashboards. API-only — pipe to Tableau, PowerBI, Looker.\n- Conversation Intelligence webhooks triggering traffic control: Must write custom Functions to act on signals.\n\nWhat does NOT work at GA:\n- AI copilot silently listening during human conversation (Conversation Orchestrator participant modes)\n- Supervisor whisper\u002Fbarge via Conversation Orchestrator (use existing Flex\u002FConference patterns)\n- Native \"Next Best Action\" auto-execution (operator suggests, human\u002Fbackend decides)\n- Automated intervention pausing outbound campaigns (planned)\n\n## Output Format\n\nAfter qualifying the developer, recommend:\n\n```\nRecommended Architecture: [Level 1-4 description]\n\nProduct Skills to Install:\n- twilio-call-recordings (if Level 1+, recording needed)\n- twilio-conversation-intelligence (if Level 2+)\n- twilio-customer-memory (if Level 3+)\n- twilio-conversation-orchestrator (if Level 3+)\n- twilio-taskrouter-routing (if Level 4)\n- twilio-voice-insights (for call quality diagnostics)\n- twilio-sendgrid-email-send (if post-call summary emails needed)\n\nSetup Skills:\n- twilio-account-setup\n- twilio-iam-auth-setup\n- twilio-webhook-architecture\n\nGuardrail Skills:\n- twilio-security-hardening (always)\n- twilio-debugging-observability (always — Voice Insights, Event Streams, error triage)\n```\n",{"data":40,"body":42},{"name":4,"description":6,"tier":41},"discover",{"type":43,"children":44},"root",[45,54,60,66,71,106,112,123,133,143,149,283,289,296,325,365,371,398,482,488,513,557,563,588,617,623,629,662,668,676,694,702,725,733,751,757,762,860,866,872,910,916,961,967,1085,1091,1096,1119,1124,1142,1147,1170,1176,1181],{"type":46,"tag":47,"props":48,"children":50},"element","h2",{"id":49},"role",[51],{"type":52,"value":53},"text","Role",{"type":46,"tag":55,"props":56,"children":57},"p",{},[58],{"type":52,"value":59},"You are a Human Agent Augmentation Advisor. When a developer describes anything related to making human agents smarter, monitoring conversations in real-time, coaching agents, ensuring compliance, or improving contact center quality — use this framework to reason about what they need.",{"type":46,"tag":47,"props":61,"children":63},{"id":62},"when-this-skill-activates",[64],{"type":52,"value":65},"When This Skill Activates",{"type":46,"tag":55,"props":67,"children":68},{},[69],{"type":52,"value":70},"Trigger on any of these signals:",{"type":46,"tag":72,"props":73,"children":74},"ul",{},[75,81,86,91,96,101],{"type":46,"tag":76,"props":77,"children":78},"li",{},[79],{"type":52,"value":80},"\"Agent assist,\" \"agent coaching,\" \"real-time coaching,\" \"agent copilot\"",{"type":46,"tag":76,"props":82,"children":83},{},[84],{"type":52,"value":85},"\"Script adherence,\" \"compliance monitoring,\" \"QA automation\"",{"type":46,"tag":76,"props":87,"children":88},{},[89],{"type":52,"value":90},"\"Sentiment detection,\" \"next best response,\" \"live prompting\"",{"type":46,"tag":76,"props":92,"children":93},{},[94],{"type":52,"value":95},"\"Call transcription,\" \"conversation analytics,\" \"call center intelligence\"",{"type":46,"tag":76,"props":97,"children":98},{},[99],{"type":52,"value":100},"\"Conversation Intelligence,\" \"Language Operators,\" \"Conversational Intelligence\"",{"type":46,"tag":76,"props":102,"children":103},{},[104],{"type":52,"value":105},"Any request to analyze, monitor, or augment live human conversations",{"type":46,"tag":47,"props":107,"children":109},{"id":108},"step-1-detect-specificity-and-decide-your-mode",[110],{"type":52,"value":111},"Step 1: Detect Specificity and Decide Your Mode",{"type":46,"tag":55,"props":113,"children":114},{},[115,121],{"type":46,"tag":116,"props":117,"children":118},"strong",{},[119],{"type":52,"value":120},"High-level request",{"type":52,"value":122}," (e.g., \"I want AI to help my agents perform better\"):\n→ DISCOVERY MODE. Walk through Steps 2-4 to understand what \"better\" means.",{"type":46,"tag":55,"props":124,"children":125},{},[126,131],{"type":46,"tag":116,"props":127,"children":128},{},[129],{"type":52,"value":130},"Mid-level request",{"type":52,"value":132}," (e.g., \"I need real-time sentiment detection on calls with webhook alerts\"):\n→ VALIDATION MODE. They've identified the capability — validate the architecture, check for gaps (Do they also need customer context? Recording for post-call?), recommend skills.",{"type":46,"tag":55,"props":134,"children":135},{},[136,141],{"type":46,"tag":116,"props":137,"children":138},{},[139],{"type":52,"value":140},"Specific implementation request",{"type":52,"value":142}," (e.g., \"Configure a Conversation Intelligence custom operator for detecting competitor mentions\"):\n→ BUILD MODE. Proceed with the relevant Product skill. Quick context check: Is Conversation Intelligence provisioned? Is Conversation Orchestrator linked? Are they aware of the operator lifecycle gotchas?",{"type":46,"tag":47,"props":144,"children":146},{"id":145},"step-2-qualify-intent-the-5-essential-questions",[147],{"type":52,"value":148},"Step 2: Qualify Intent — The 5 Essential Questions",{"type":46,"tag":150,"props":151,"children":152},"ol",{},[153,184,210,236,257],{"type":46,"tag":76,"props":154,"children":155},{},[156,161],{"type":46,"tag":116,"props":157,"children":158},{},[159],{"type":52,"value":160},"What does \"augmentation\" mean for your agents?",{"type":46,"tag":72,"props":162,"children":163},{},[164,169,174,179],{"type":46,"tag":76,"props":165,"children":166},{},[167],{"type":52,"value":168},"Real-time coaching: Live suggestions\u002Fprompts appearing on the agent's screen during a call",{"type":46,"tag":76,"props":170,"children":171},{},[172],{"type":52,"value":173},"Compliance monitoring: Automated detection of script deviations, regulatory violations, disclosure requirements",{"type":46,"tag":76,"props":175,"children":176},{},[177],{"type":52,"value":178},"Post-call QA: Automated scoring and review of completed conversations (replacing manual sampling)",{"type":46,"tag":76,"props":180,"children":181},{},[182],{"type":52,"value":183},"Intelligent routing: Using AI signals to send calls to the right specialist",{"type":46,"tag":76,"props":185,"children":186},{},[187,192],{"type":46,"tag":116,"props":188,"children":189},{},[190],{"type":52,"value":191},"What channels are your agents handling?",{"type":46,"tag":72,"props":193,"children":194},{},[195,200,205],{"type":46,"tag":76,"props":196,"children":197},{},[198],{"type":52,"value":199},"Voice calls only → Transcription + Conversation Intelligence operators on audio stream",{"type":46,"tag":76,"props":201,"children":202},{},[203],{"type":52,"value":204},"Voice + messaging → Conversation Orchestrator for unified conversation tracking + Conversation Intelligence across both",{"type":46,"tag":76,"props":206,"children":207},{},[208],{"type":52,"value":209},"Messaging only → Conversation Intelligence operators on text (no transcription needed)",{"type":46,"tag":76,"props":211,"children":212},{},[213,218],{"type":46,"tag":116,"props":214,"children":215},{},[216],{"type":52,"value":217},"What's your existing contact center infrastructure?",{"type":46,"tag":72,"props":219,"children":220},{},[221,226,231],{"type":46,"tag":76,"props":222,"children":223},{},[224],{"type":52,"value":225},"Twilio Flex → Native integration path (Flex Agent Copilot replatforming onto Conversation Intelligence)",{"type":46,"tag":76,"props":227,"children":228},{},[229],{"type":52,"value":230},"Other CCaaS (Genesys, Five9, NICE) → Webhook-based integration, more custom glue",{"type":46,"tag":76,"props":232,"children":233},{},[234],{"type":52,"value":235},"Custom-built → Full flexibility but more setup",{"type":46,"tag":76,"props":237,"children":238},{},[239,244],{"type":46,"tag":116,"props":240,"children":241},{},[242],{"type":52,"value":243},"Do you need customer context surfaced to agents?",{"type":46,"tag":72,"props":245,"children":246},{},[247,252],{"type":46,"tag":76,"props":248,"children":249},{},[250],{"type":52,"value":251},"No (agents look up context themselves) → Skip Conversation Memory",{"type":46,"tag":76,"props":253,"children":254},{},[255],{"type":52,"value":256},"Yes (show customer history, preferences, past issues on accept) → Add Conversation Memory",{"type":46,"tag":76,"props":258,"children":259},{},[260,265],{"type":46,"tag":116,"props":261,"children":262},{},[263],{"type":52,"value":264},"What's your call volume and budget sensitivity?",{"type":46,"tag":72,"props":266,"children":267},{},[268,273,278],{"type":46,"tag":76,"props":269,"children":270},{},[271],{"type":52,"value":272},"Not all calls are worth transcribing",{"type":46,"tag":76,"props":274,"children":275},{},[276],{"type":52,"value":277},"Consider selective intelligence: Apply Conversation Intelligence only to specific queues, customer segments, or call types",{"type":46,"tag":76,"props":279,"children":280},{},[281],{"type":52,"value":282},"Conversation Intelligence pricing is per-conversation-character — model selection affects cost (GPT-4.1-nano for speed\u002Fcost vs. GPT-5.2 for quality)",{"type":46,"tag":47,"props":284,"children":286},{"id":285},"step-3-assess-sophistication-the-capability-ladder",[287],{"type":52,"value":288},"Step 3: Assess Sophistication — The Capability Ladder",{"type":46,"tag":290,"props":291,"children":293},"h3",{"id":292},"level-1-listen-transcription-recording",[294],{"type":52,"value":295},"Level 1: Listen — Transcription & Recording",{"type":46,"tag":55,"props":297,"children":298},{},[299,304,306,311,313,318,320],{"type":46,"tag":116,"props":300,"children":301},{},[302],{"type":52,"value":303},"Developer says:",{"type":52,"value":305}," \"I want to transcribe calls for review and analysis.\"\n",{"type":46,"tag":116,"props":307,"children":308},{},[309],{"type":52,"value":310},"Architecture:",{"type":52,"value":312}," Real-time Transcription + Call Recordings\n",{"type":46,"tag":116,"props":314,"children":315},{},[316],{"type":52,"value":317},"What it does:",{"type":52,"value":319}," Live STT during calls → transcripts available for search and review. Recordings stored for compliance and playback.\n",{"type":46,"tag":116,"props":321,"children":322},{},[323],{"type":52,"value":324},"Key decisions:",{"type":46,"tag":72,"props":326,"children":327},{},[328,333,338],{"type":46,"tag":76,"props":329,"children":330},{},[331],{"type":52,"value":332},"Engine: Google (wider language support) vs Deepgram (better accuracy, lower latency)",{"type":46,"tag":76,"props":334,"children":335},{},[336],{"type":52,"value":337},"Track: Inbound audio, outbound audio, or both",{"type":46,"tag":76,"props":339,"children":340},{},[341,343,350,352,357,359],{"type":52,"value":342},"Recording method: ",{"type":46,"tag":344,"props":345,"children":347},"code",{"className":346},[],[348],{"type":52,"value":349},"\u003CDial record=\"record-from-answer\">",{"type":52,"value":351}," for simplicity, or Recordings REST API for control\n",{"type":46,"tag":116,"props":353,"children":354},{},[355],{"type":52,"value":356},"Skills to install:",{"type":52,"value":358}," ",{"type":46,"tag":344,"props":360,"children":362},{"className":361},[],[363],{"type":52,"value":364},"twilio-call-recordings",{"type":46,"tag":290,"props":366,"children":368},{"id":367},"level-2-coach-real-time-intelligence",[369],{"type":52,"value":370},"Level 2: Coach — Real-Time Intelligence",{"type":46,"tag":55,"props":372,"children":373},{},[374,378,380,384,386,391,393],{"type":46,"tag":116,"props":375,"children":376},{},[377],{"type":52,"value":303},{"type":52,"value":379}," \"I want to detect sentiment, prompt agents with next-best-response, or monitor script adherence live.\"\n",{"type":46,"tag":116,"props":381,"children":382},{},[383],{"type":52,"value":310},{"type":52,"value":385}," Level 1 + Conversation Intelligence v3 Language Operators\n",{"type":46,"tag":116,"props":387,"children":388},{},[389],{"type":52,"value":390},"What it adds:",{"type":52,"value":392}," Conversation Intelligence attaches to live conversations → runs operators in parallel → fires webhooks on signal detection → your backend pushes prompts to agent UI\n",{"type":46,"tag":116,"props":394,"children":395},{},[396],{"type":52,"value":397},"Pre-built operators (GA):",{"type":46,"tag":72,"props":399,"children":400},{},[401,411,421,431,441,455,460,465],{"type":46,"tag":76,"props":402,"children":403},{},[404,409],{"type":46,"tag":116,"props":405,"children":406},{},[407],{"type":52,"value":408},"Sentiment:",{"type":52,"value":410}," Detect caller frustration, anger, satisfaction in real-time",{"type":46,"tag":76,"props":412,"children":413},{},[414,419],{"type":46,"tag":116,"props":415,"children":416},{},[417],{"type":52,"value":418},"Script Adherence:",{"type":52,"value":420}," Flag when agent deviates from required script (compliance disclosures, greeting, etc.)",{"type":46,"tag":76,"props":422,"children":423},{},[424,429],{"type":46,"tag":116,"props":425,"children":426},{},[427],{"type":52,"value":428},"Next Best Response (NBR):",{"type":52,"value":430}," Suggest the best reply based on conversation context",{"type":46,"tag":76,"props":432,"children":433},{},[434,439],{"type":46,"tag":116,"props":435,"children":436},{},[437],{"type":52,"value":438},"Summary:",{"type":52,"value":440}," Auto-generate post-call summaries",{"type":46,"tag":76,"props":442,"children":443},{},[444,449,451],{"type":46,"tag":116,"props":445,"children":446},{},[447],{"type":52,"value":448},"Custom Operators:",{"type":52,"value":450}," Define your own detection rules (competitor mentions, churn signals, upsell opportunities)\n",{"type":46,"tag":116,"props":452,"children":453},{},[454],{"type":52,"value":324},{"type":46,"tag":76,"props":456,"children":457},{},[458],{"type":52,"value":459},"Which operators to activate (each adds latency and cost)",{"type":46,"tag":76,"props":461,"children":462},{},[463],{"type":52,"value":464},"Webhook destination: Where do signals go? (Flex plugin, custom dashboard, Slack alert)",{"type":46,"tag":76,"props":466,"children":467},{},[468,470,474,476],{"type":52,"value":469},"Model profile: Speed (GPT-4.1-nano, lower cost) vs quality (GPT-5.2, higher accuracy)\n",{"type":46,"tag":116,"props":471,"children":472},{},[473],{"type":52,"value":356},{"type":52,"value":475}," + ",{"type":46,"tag":344,"props":477,"children":479},{"className":478},[],[480],{"type":52,"value":481},"twilio-conversation-intelligence",{"type":46,"tag":290,"props":483,"children":485},{"id":484},"level-3-context-customer-memory-for-agents",[486],{"type":52,"value":487},"Level 3: Context — Customer Memory for Agents",{"type":46,"tag":55,"props":489,"children":490},{},[491,495,497,501,503,507,509],{"type":46,"tag":116,"props":492,"children":493},{},[494],{"type":52,"value":303},{"type":52,"value":496}," \"When the agent picks up, I want them to see who this customer is and their full history.\"\n",{"type":46,"tag":116,"props":498,"children":499},{},[500],{"type":52,"value":310},{"type":52,"value":502}," Level 2 + Conversation Memory (profile hydration)\n",{"type":46,"tag":116,"props":504,"children":505},{},[506],{"type":52,"value":390},{"type":52,"value":508}," On task acceptance, agent desktop fetches Conversation Memory profile → displays customer summary, traits, past observations → agent starts the conversation with full context instead of \"Who is this? What do you need?\"\n",{"type":46,"tag":116,"props":510,"children":511},{},[512],{"type":52,"value":324},{"type":46,"tag":72,"props":514,"children":515},{},[516,521,526],{"type":46,"tag":76,"props":517,"children":518},{},[519],{"type":52,"value":520},"What to surface: Summary only (GA for Flex) or deep context (traits, recent observations, Segment data)",{"type":46,"tag":76,"props":522,"children":523},{},[524],{"type":52,"value":525},"Identity resolution: Match incoming caller to Conversation Memory profile by phone number, email, or custom ID",{"type":46,"tag":76,"props":527,"children":528},{},[529,531,536,538,542,543,549,551],{"type":52,"value":530},"Enrichment sources: Conversation Memory observations only, or also Segment traits via Bridge\n",{"type":46,"tag":116,"props":532,"children":533},{},[534],{"type":52,"value":535},"GA constraint:",{"type":52,"value":537}," Flex integration is summary-only at GA. Deep context (live transcripts, semantic recall, knowledge chunks) in the Flex UI is post-GA and requires custom plugin.\n",{"type":46,"tag":116,"props":539,"children":540},{},[541],{"type":52,"value":356},{"type":52,"value":475},{"type":46,"tag":344,"props":544,"children":546},{"className":545},[],[547],{"type":52,"value":548},"twilio-customer-memory",{"type":52,"value":550},", ",{"type":46,"tag":344,"props":552,"children":554},{"className":553},[],[555],{"type":52,"value":556},"twilio-conversation-orchestrator",{"type":46,"tag":290,"props":558,"children":560},{"id":559},"level-4-route-intelligence-driven-routing",[561],{"type":52,"value":562},"Level 4: Route — Intelligence-Driven Routing",{"type":46,"tag":55,"props":564,"children":565},{},[566,570,572,576,578,582,584],{"type":46,"tag":116,"props":567,"children":568},{},[569],{"type":52,"value":303},{"type":52,"value":571}," \"I want AI signals to determine which agent gets the call — not just FIFO.\"\n",{"type":46,"tag":116,"props":573,"children":574},{},[575],{"type":52,"value":310},{"type":52,"value":577}," Level 3 + TaskRouter consuming Conversation Intelligence signals\n",{"type":46,"tag":116,"props":579,"children":580},{},[581],{"type":52,"value":390},{"type":52,"value":583}," Conversation Intelligence emits structured routing signals (intent, sentiment, skill_needed, VIP detection) → these feed into TaskRouter workflow expressions → calls route to specialized skill groups (retention team, technical support, VIP desk)\n",{"type":46,"tag":116,"props":585,"children":586},{},[587],{"type":52,"value":324},{"type":46,"tag":72,"props":589,"children":590},{},[591,596,601],{"type":46,"tag":76,"props":592,"children":593},{},[594],{"type":52,"value":595},"Which Conversation Intelligence signals feed routing? (intent classification, sentiment threshold, customer segment from Conversation Memory)",{"type":46,"tag":76,"props":597,"children":598},{},[599],{"type":52,"value":600},"TaskRouter workflow design: Simple skills-matching or multi-tier escalation",{"type":46,"tag":76,"props":602,"children":603},{},[604,606,610,611],{"type":52,"value":605},"Overflow strategy: What happens when the target queue is full?\n",{"type":46,"tag":116,"props":607,"children":608},{},[609],{"type":52,"value":356},{"type":52,"value":475},{"type":46,"tag":344,"props":612,"children":614},{"className":613},[],[615],{"type":52,"value":616},"twilio-taskrouter-routing",{"type":46,"tag":47,"props":618,"children":620},{"id":619},"step-4-qualify-context",[621],{"type":52,"value":622},"Step 4: Qualify Context",{"type":46,"tag":290,"props":624,"children":626},{"id":625},"existing-infrastructure",[627],{"type":52,"value":628},"Existing Infrastructure",{"type":46,"tag":72,"props":630,"children":631},{},[632,642,652],{"type":46,"tag":76,"props":633,"children":634},{},[635,640],{"type":46,"tag":116,"props":636,"children":637},{},[638],{"type":52,"value":639},"Flex customer:",{"type":52,"value":641}," Leverage Flex Agent Copilot (being replatformed onto Conversation Intelligence). Tightest integration path.",{"type":46,"tag":76,"props":643,"children":644},{},[645,650],{"type":46,"tag":116,"props":646,"children":647},{},[648],{"type":52,"value":649},"Other CCaaS:",{"type":52,"value":651}," You'll integrate via webhooks. Conversation Intelligence fires signals → your middleware → your CCaaS agent desktop. More work but fully functional.",{"type":46,"tag":76,"props":653,"children":654},{},[655,660],{"type":46,"tag":116,"props":656,"children":657},{},[658],{"type":52,"value":659},"No contact center yet:",{"type":52,"value":661}," Consider starting with Flex + TaskRouter as the foundation, then layer intelligence.",{"type":46,"tag":290,"props":663,"children":665},{"id":664},"customer-profile",[666],{"type":52,"value":667},"Customer Profile",{"type":46,"tag":55,"props":669,"children":670},{},[671],{"type":46,"tag":116,"props":672,"children":673},{},[674],{"type":52,"value":675},"ISV (building augmentation for multiple clients):",{"type":46,"tag":72,"props":677,"children":678},{},[679,684,689],{"type":46,"tag":76,"props":680,"children":681},{},[682],{"type":52,"value":683},"Per-client Conversation Intelligence operator configurations",{"type":46,"tag":76,"props":685,"children":686},{},[687],{"type":52,"value":688},"Separate Conversation Memory stores per client (max 15 per account)",{"type":46,"tag":76,"props":690,"children":691},{},[692],{"type":52,"value":693},"White-label considerations for agent UI",{"type":46,"tag":55,"props":695,"children":696},{},[697],{"type":46,"tag":116,"props":698,"children":699},{},[700],{"type":52,"value":701},"Enterprise:",{"type":46,"tag":72,"props":703,"children":704},{},[705,710,715,720],{"type":46,"tag":76,"props":706,"children":707},{},[708],{"type":52,"value":709},"Compliance operators are likely mandatory (regulated industries: finance, healthcare, insurance)",{"type":46,"tag":76,"props":711,"children":712},{},[713],{"type":52,"value":714},"Selective intelligence to control cost at scale",{"type":46,"tag":76,"props":716,"children":717},{},[718],{"type":52,"value":719},"Integration with existing QA workflows (Calabrio, Verint, etc.)",{"type":46,"tag":76,"props":721,"children":722},{},[723],{"type":52,"value":724},"No ngrok for webhook delivery — deploy to production infrastructure",{"type":46,"tag":55,"props":726,"children":727},{},[728],{"type":46,"tag":116,"props":729,"children":730},{},[731],{"type":52,"value":732},"SMB:",{"type":46,"tag":72,"props":734,"children":735},{},[736,741,746],{"type":46,"tag":76,"props":737,"children":738},{},[739],{"type":52,"value":740},"Start at Level 2 — sentiment + summary operators give immediate value",{"type":46,"tag":76,"props":742,"children":743},{},[744],{"type":52,"value":745},"Skip Conversation Memory initially — add when agent \"amnesia\" becomes a pain point",{"type":46,"tag":76,"props":747,"children":748},{},[749],{"type":52,"value":750},"Use pre-built operators before investing in custom ones",{"type":46,"tag":47,"props":752,"children":754},{"id":753},"architectural-warnings",[755],{"type":52,"value":756},"Architectural Warnings",{"type":46,"tag":55,"props":758,"children":759},{},[760],{"type":52,"value":761},"These affect which capabilities to recommend and how to set expectations — implementation details are in the Product skills.",{"type":46,"tag":72,"props":763,"children":764},{},[765,775,785,802,820,830,840,850],{"type":46,"tag":76,"props":766,"children":767},{},[768,773],{"type":46,"tag":116,"props":769,"children":770},{},[771],{"type":52,"value":772},"Silent linkage chain:",{"type":52,"value":774}," Conversations Service → Intelligence Service → Capture Rules → Operators must be linked in sequence. Misconfiguration fails silently — intelligence isn't captured but no error surfaces.",{"type":46,"tag":76,"props":776,"children":777},{},[778,783],{"type":46,"tag":116,"props":779,"children":780},{},[781],{"type":52,"value":782},"Operator lifecycle trap:",{"type":52,"value":784}," PUT on an operator creates an inactive new version. No activation endpoint exists — must delete and POST a new one. Plan operator changes as delete+recreate, not update.",{"type":46,"tag":76,"props":786,"children":787},{},[788,793,794,800],{"type":46,"tag":116,"props":789,"children":790},{},[791],{"type":52,"value":792},"One-way door settings:",{"type":52,"value":358},{"type":46,"tag":344,"props":795,"children":797},{"className":796},[],[798],{"type":52,"value":799},"GROUP_BY_PARTICIPANT_ADDRESSES",{"type":52,"value":801}," on a Conversations Service is immutable once set. Removing a capture rule stops ALL capture for that service.",{"type":46,"tag":76,"props":803,"children":804},{},[805,810,812,818],{"type":46,"tag":116,"props":806,"children":807},{},[808],{"type":52,"value":809},"OperatorResults scope leak:",{"type":52,"value":811}," API may return results from other conversations on the same account. Always filter by ",{"type":46,"tag":344,"props":813,"children":815},{"className":814},[],[816],{"type":52,"value":817},"conversation_id",{"type":52,"value":819},".",{"type":46,"tag":76,"props":821,"children":822},{},[823,828],{"type":46,"tag":116,"props":824,"children":825},{},[826],{"type":52,"value":827},"Dashboard vs. webhooks:",{"type":52,"value":829}," Conversation Intelligence signals take 7-10 minutes to reach the dashboard. For real-time coaching, rely on webhook delivery — not dashboard polling.",{"type":46,"tag":76,"props":831,"children":832},{},[833,838],{"type":46,"tag":116,"props":834,"children":835},{},[836],{"type":52,"value":837},"Flex GA constraint:",{"type":52,"value":839}," Conversation Memory integration in Flex is summary-only at GA. Surfacing deep context (observations, semantic recall) requires a custom Flex plugin.",{"type":46,"tag":76,"props":841,"children":842},{},[843,848],{"type":46,"tag":116,"props":844,"children":845},{},[846],{"type":52,"value":847},"Cost model:",{"type":52,"value":849}," Conversation Intelligence pricing is per-conversation-character. Model selection (GPT-4.1-nano for speed\u002Fcost vs. GPT-5.2 for quality) directly affects bill. Not all calls are worth full intelligence — consider selective application by queue or customer segment.",{"type":46,"tag":76,"props":851,"children":852},{},[853,858],{"type":46,"tag":116,"props":854,"children":855},{},[856],{"type":52,"value":857},"No SDK at GA:",{"type":52,"value":859}," All Twilio Conversations integration is raw HTTP with Basic Auth. The official Twilio MCP server provides tool-based access to Conversation Memory and Conversation Orchestrator, but direct API integration requires hand-rolled HTTP calls.",{"type":46,"tag":47,"props":861,"children":863},{"id":862},"decision-rules",[864],{"type":52,"value":865},"Decision Rules",{"type":46,"tag":290,"props":867,"children":869},{"id":868},"transcription-engine-selection",[870],{"type":52,"value":871},"Transcription Engine Selection",{"type":46,"tag":72,"props":873,"children":874},{},[875,885,895,905],{"type":46,"tag":76,"props":876,"children":877},{},[878,883],{"type":46,"tag":116,"props":879,"children":880},{},[881],{"type":52,"value":882},"Google STT:",{"type":52,"value":884}," Wider language support, good for international contact centers. Choose when multi-lingual support is the priority.",{"type":46,"tag":76,"props":886,"children":887},{},[888,893],{"type":46,"tag":116,"props":889,"children":890},{},[891],{"type":52,"value":892},"Deepgram:",{"type":52,"value":894}," Lower latency, better accuracy for English. Choose for English-primary contact centers or noisy environments.",{"type":46,"tag":76,"props":896,"children":897},{},[898,903],{"type":46,"tag":116,"props":899,"children":900},{},[901],{"type":52,"value":902},"Dual-track recommended:",{"type":52,"value":904}," Enables speaker diarization — Conversation Intelligence can distinguish agent from caller. Single-track reduces script adherence and sentiment accuracy.",{"type":46,"tag":76,"props":906,"children":907},{},[908],{"type":52,"value":909},"Implementation gotchas: callback format, ordering, short utterances — see Twilio Real-Time Transcription docs.",{"type":46,"tag":290,"props":911,"children":913},{"id":912},"conversation-intelligence-operator-selection",[914],{"type":52,"value":915},"Conversation Intelligence Operator Selection",{"type":46,"tag":72,"props":917,"children":918},{},[919,929,939,949],{"type":46,"tag":76,"props":920,"children":921},{},[922,927],{"type":46,"tag":116,"props":923,"children":924},{},[925],{"type":52,"value":926},"Pre-built operators:",{"type":52,"value":928}," Sentiment, Script Adherence, Next Best Response, Summary. Start here — immediate value, no custom configuration.",{"type":46,"tag":76,"props":930,"children":931},{},[932,937],{"type":46,"tag":116,"props":933,"children":934},{},[935],{"type":52,"value":936},"Custom operators:",{"type":52,"value":938}," For domain-specific detection (competitor mentions, churn signals, upsell opportunities). Three types: text-generation, classification, extraction.",{"type":46,"tag":76,"props":940,"children":941},{},[942,947],{"type":46,"tag":116,"props":943,"children":944},{},[945],{"type":52,"value":946},"Selective application:",{"type":52,"value":948}," Not all calls warrant full intelligence. Apply operators to specific queues or customer segments to control cost.",{"type":46,"tag":76,"props":950,"children":951},{},[952,954,959],{"type":52,"value":953},"Operator lifecycle gotchas (PUT trap, capture rule deletion) are documented in the ",{"type":46,"tag":344,"props":955,"children":957},{"className":956},[],[958],{"type":52,"value":481},{"type":52,"value":960}," skill.",{"type":46,"tag":290,"props":962,"children":964},{"id":963},"recording-method-selection",[965],{"type":52,"value":966},"Recording Method Selection",{"type":46,"tag":72,"props":968,"children":969},{},[970,988,998,1022,1039,1056,1074],{"type":46,"tag":76,"props":971,"children":972},{},[973,986],{"type":46,"tag":116,"props":974,"children":975},{},[976,978,984],{"type":52,"value":977},"Use ",{"type":46,"tag":344,"props":979,"children":981},{"className":980},[],[982],{"type":52,"value":983},"\u003CDial record>",{"type":52,"value":985}," when:",{"type":52,"value":987}," Simple two-party call recording. Minimal setup.",{"type":46,"tag":76,"props":989,"children":990},{},[991,996],{"type":46,"tag":116,"props":992,"children":993},{},[994],{"type":52,"value":995},"Use Recordings REST API when:",{"type":52,"value":997}," Mid-call control needed (pause during payment). Dual-channel recording for QA.",{"type":46,"tag":76,"props":999,"children":1000},{},[1001,1012,1014,1020],{"type":46,"tag":116,"props":1002,"children":1003},{},[1004,1005,1011],{"type":52,"value":977},{"type":46,"tag":344,"props":1006,"children":1008},{"className":1007},[],[1009],{"type":52,"value":1010},"\u003CStart>\u003CRecording>",{"type":52,"value":985},{"type":52,"value":1013}," Recording must start before ",{"type":46,"tag":344,"props":1015,"children":1017},{"className":1016},[],[1018],{"type":52,"value":1019},"\u003CConnect>",{"type":52,"value":1021}," (e.g., ConversationRelay AI side).",{"type":46,"tag":76,"props":1023,"children":1024},{},[1025,1037],{"type":46,"tag":116,"props":1026,"children":1027},{},[1028,1030,1036],{"type":52,"value":1029},"Use Conference ",{"type":46,"tag":344,"props":1031,"children":1033},{"className":1032},[],[1034],{"type":52,"value":1035},"record",{"type":52,"value":985},{"type":52,"value":1038}," Multi-party calls.",{"type":46,"tag":76,"props":1040,"children":1041},{},[1042,1047,1048,1054],{"type":46,"tag":116,"props":1043,"children":1044},{},[1045],{"type":52,"value":1046},"Critical:",{"type":52,"value":358},{"type":46,"tag":344,"props":1049,"children":1051},{"className":1050},[],[1052],{"type":52,"value":1053},"\u003CRecord>",{"type":52,"value":1055}," (standalone verb) is voicemail-style — NOT for recording calls.",{"type":46,"tag":76,"props":1057,"children":1058},{},[1059,1064,1066,1072],{"type":46,"tag":116,"props":1060,"children":1061},{},[1062],{"type":52,"value":1063},"PCI:",{"type":52,"value":1065}," Never record card numbers. Use ",{"type":46,"tag":344,"props":1067,"children":1069},{"className":1068},[],[1070],{"type":52,"value":1071},"\u003CPay>",{"type":52,"value":1073}," verb. PCI Mode is IRREVERSIBLE and account-wide.",{"type":46,"tag":76,"props":1075,"children":1076},{},[1077,1079,1084],{"type":52,"value":1078},"Detailed method comparison and gotchas are in the ",{"type":46,"tag":344,"props":1080,"children":1082},{"className":1081},[],[1083],{"type":52,"value":364},{"type":52,"value":960},{"type":46,"tag":47,"props":1086,"children":1088},{"id":1087},"ga-constraints-may-2026",[1089],{"type":52,"value":1090},"GA Constraints (May 2026)",{"type":46,"tag":55,"props":1092,"children":1093},{},[1094],{"type":52,"value":1095},"What works:",{"type":46,"tag":72,"props":1097,"children":1098},{},[1099,1104,1109,1114],{"type":46,"tag":76,"props":1100,"children":1101},{},[1102],{"type":52,"value":1103},"Conversation Intelligence v3 real-time operators (sentiment, script adherence, NBR, custom) ✅",{"type":46,"tag":76,"props":1105,"children":1106},{},[1107],{"type":52,"value":1108},"Conversation Memory profile storage and Recall ✅",{"type":46,"tag":76,"props":1110,"children":1111},{},[1112],{"type":52,"value":1113},"TaskRouter with custom routing signals ✅",{"type":46,"tag":76,"props":1115,"children":1116},{},[1117],{"type":52,"value":1118},"Call recordings and real-time transcription ✅",{"type":46,"tag":55,"props":1120,"children":1121},{},[1122],{"type":52,"value":1123},"What requires custom code:",{"type":46,"tag":72,"props":1125,"children":1126},{},[1127,1132,1137],{"type":46,"tag":76,"props":1128,"children":1129},{},[1130],{"type":52,"value":1131},"Flex Agent Copilot: Being replatformed onto Conversation Intelligence. Early stages — expect custom plugin work.",{"type":46,"tag":76,"props":1133,"children":1134},{},[1135],{"type":52,"value":1136},"Aggregated insights: No native dashboards. API-only — pipe to Tableau, PowerBI, Looker.",{"type":46,"tag":76,"props":1138,"children":1139},{},[1140],{"type":52,"value":1141},"Conversation Intelligence webhooks triggering traffic control: Must write custom Functions to act on signals.",{"type":46,"tag":55,"props":1143,"children":1144},{},[1145],{"type":52,"value":1146},"What does NOT work at GA:",{"type":46,"tag":72,"props":1148,"children":1149},{},[1150,1155,1160,1165],{"type":46,"tag":76,"props":1151,"children":1152},{},[1153],{"type":52,"value":1154},"AI copilot silently listening during human conversation (Conversation Orchestrator participant modes)",{"type":46,"tag":76,"props":1156,"children":1157},{},[1158],{"type":52,"value":1159},"Supervisor whisper\u002Fbarge via Conversation Orchestrator (use existing Flex\u002FConference patterns)",{"type":46,"tag":76,"props":1161,"children":1162},{},[1163],{"type":52,"value":1164},"Native \"Next Best Action\" auto-execution (operator suggests, human\u002Fbackend decides)",{"type":46,"tag":76,"props":1166,"children":1167},{},[1168],{"type":52,"value":1169},"Automated intervention pausing outbound campaigns (planned)",{"type":46,"tag":47,"props":1171,"children":1173},{"id":1172},"output-format",[1174],{"type":52,"value":1175},"Output Format",{"type":46,"tag":55,"props":1177,"children":1178},{},[1179],{"type":52,"value":1180},"After qualifying the developer, recommend:",{"type":46,"tag":1182,"props":1183,"children":1187},"pre",{"className":1184,"code":1186,"language":52},[1185],"language-text","Recommended Architecture: [Level 1-4 description]\n\nProduct Skills to Install:\n- twilio-call-recordings (if Level 1+, recording needed)\n- twilio-conversation-intelligence (if Level 2+)\n- twilio-customer-memory (if Level 3+)\n- twilio-conversation-orchestrator (if Level 3+)\n- twilio-taskrouter-routing (if Level 4)\n- twilio-voice-insights (for call quality diagnostics)\n- twilio-sendgrid-email-send (if post-call summary emails needed)\n\nSetup Skills:\n- twilio-account-setup\n- twilio-iam-auth-setup\n- twilio-webhook-architecture\n\nGuardrail Skills:\n- twilio-security-hardening (always)\n- twilio-debugging-observability (always — Voice Insights, Event Streams, error triage)\n",[1188],{"type":46,"tag":344,"props":1189,"children":1191},{"__ignoreMap":1190},"",[1192],{"type":52,"value":1186},{"items":1194,"total":1313},[1195,1213,1227,1239,1259,1281,1301],{"slug":1196,"name":1196,"fn":1197,"description":1198,"org":1199,"tags":1200,"stars":28,"repoUrl":29,"updatedAt":30},"accessibility-and-inclusive-visualization","make data visualizations accessible","Make data visualizations accessible and inclusive. Use when the user needs chart or diagram accessibility guidance, text alternatives for complex visuals, color and contrast review, keyboard support, reduced-motion behavior for animation or parallax, or an accessibility QA workflow for exported figures, UML-like diagrams, and dashboards.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[1201,1204,1207,1210],{"name":1202,"slug":1203,"type":15},"Accessibility","accessibility",{"name":1205,"slug":1206,"type":15},"Charts","charts",{"name":1208,"slug":1209,"type":15},"Data Visualization","data-visualization",{"name":1211,"slug":1212,"type":15},"Design","design",{"slug":1214,"name":1214,"fn":1215,"description":1216,"org":1217,"tags":1218,"stars":28,"repoUrl":29,"updatedAt":1226},"agent-browser","automate browser interactions for agents","Browser automation CLI for AI agents. Use when the user needs to interact with websites, verify dev server output, test web apps, navigate pages, fill forms, click buttons, take screenshots, extract data, or automate any browser task. Also triggers when a dev server starts so you can verify it visually.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[1219,1220,1223],{"name":20,"slug":21,"type":15},{"name":1221,"slug":1222,"type":15},"Browser Automation","browser-automation",{"name":1224,"slug":1225,"type":15},"Testing","testing","2026-04-06T18:41:03.44016",{"slug":1228,"name":1228,"fn":1229,"description":1230,"org":1231,"tags":1232,"stars":28,"repoUrl":29,"updatedAt":1238},"agent-browser-verify","verify dev server output with automated browser","Automated browser verification for dev servers. Triggers when a dev server starts to run a visual gut-check with agent-browser — verifies the page loads, checks for console errors, validates key UI elements, and reports pass\u002Ffail before continuing.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[1233,1234,1237],{"name":1221,"slug":1222,"type":15},{"name":1235,"slug":1236,"type":15},"Local Development","local-development",{"name":1224,"slug":1225,"type":15},"2026-04-06T18:41:17.526867",{"slug":1240,"name":1240,"fn":1241,"description":1242,"org":1243,"tags":1244,"stars":28,"repoUrl":29,"updatedAt":1258},"agents-sdk","build AI agents on Cloudflare Workers","Build AI agents on Cloudflare Workers using the Agents SDK. Load when creating stateful agents, durable workflows, real-time WebSocket apps, scheduled tasks, MCP servers, or chat applications. Covers Agent class, state management, callable RPC, Workflows integration, and React hooks. Biases towards retrieval from Cloudflare docs over pre-trained knowledge.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[1245,1246,1249,1252,1255],{"name":20,"slug":21,"type":15},{"name":1247,"slug":1248,"type":15},"Cloudflare Workers","cloudflare-workers",{"name":1250,"slug":1251,"type":15},"SDK","sdk",{"name":1253,"slug":1254,"type":15},"Serverless","serverless",{"name":1256,"slug":1257,"type":15},"WebSockets","websockets","2026-04-06T18:39:51.717063",{"slug":1260,"name":1260,"fn":1261,"description":1262,"org":1263,"tags":1264,"stars":28,"repoUrl":29,"updatedAt":1280},"ai-elements","build chat UIs with AI Elements","AI Elements component library guidance — pre-built React components for AI interfaces built on shadcn\u002Fui. Use when building chat UIs, message displays, tool call rendering, streaming responses, reasoning panels, or any AI-native interface with the AI SDK.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[1265,1268,1271,1274,1277],{"name":1266,"slug":1267,"type":15},"Frontend","frontend",{"name":1269,"slug":1270,"type":15},"React","react",{"name":1272,"slug":1273,"type":15},"shadcn\u002Fui","shadcn-ui",{"name":1275,"slug":1276,"type":15},"UI Components","ui-components",{"name":1278,"slug":1279,"type":15},"Vercel","vercel","2026-04-06T18:40:59.619419",{"slug":1282,"name":1282,"fn":1283,"description":1284,"org":1285,"tags":1286,"stars":28,"repoUrl":29,"updatedAt":1300},"ai-gateway","configure Vercel AI Gateway","Vercel AI Gateway expert guidance. Use when configuring model routing, provider failover, cost tracking, or managing multiple AI providers through a unified API.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[1287,1290,1293,1296,1299],{"name":1288,"slug":1289,"type":15},"AI Infrastructure","ai-infrastructure",{"name":1291,"slug":1292,"type":15},"Cost Optimization","cost-optimization",{"name":1294,"slug":1295,"type":15},"LLM","llm",{"name":1297,"slug":1298,"type":15},"Performance","performance",{"name":1278,"slug":1279,"type":15},"2026-04-06T18:40:44.377464",{"slug":1302,"name":1302,"fn":1303,"description":1304,"org":1305,"tags":1306,"stars":28,"repoUrl":29,"updatedAt":1312},"ai-generation-persistence","implement persistence patterns for AI generations","AI generation persistence patterns — unique IDs, addressable URLs, database storage, and cost tracking for every LLM generation",{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[1307,1308,1311],{"name":1291,"slug":1292,"type":15},{"name":1309,"slug":1310,"type":15},"Database","database",{"name":1294,"slug":1295,"type":15},"2026-04-06T18:41:08.513425",600,{"items":1315,"total":1512},[1316,1337,1360,1377,1393,1410,1429,1441,1455,1469,1481,1496],{"slug":1317,"name":1317,"fn":1318,"description":1319,"org":1320,"tags":1321,"stars":1334,"repoUrl":1335,"updatedAt":1336},"prior-auth-packet-builder","build healthcare prior authorization packets","Build a concise prior authorization packet from local case files and payer policy docs.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[1322,1325,1328,1331],{"name":1323,"slug":1324,"type":15},"Documents","documents",{"name":1326,"slug":1327,"type":15},"Healthcare","healthcare",{"name":1329,"slug":1330,"type":15},"Insurance","insurance",{"name":1332,"slug":1333,"type":15},"Regulatory Compliance","regulatory-compliance",28169,"https:\u002F\u002Fgithub.com\u002Fopenai\u002Fopenai-agents-python","2026-04-16T05:11:39.180399",{"slug":1338,"name":1338,"fn":1339,"description":1340,"org":1341,"tags":1342,"stars":1357,"repoUrl":1358,"updatedAt":1359},"aspnet-core","build ASP.NET Core web applications","Build, review, refactor, or architect ASP.NET Core web applications using current official guidance for .NET web development. Use when working on Blazor Web Apps, Razor Pages, MVC, Minimal APIs, controller-based Web APIs, SignalR, gRPC, middleware, dependency injection, configuration, authentication, authorization, testing, performance, deployment, or ASP.NET Core upgrades.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[1343,1346,1348,1351,1354],{"name":1344,"slug":1345,"type":15},".NET","dotnet",{"name":1347,"slug":1338,"type":15},"ASP.NET Core",{"name":1349,"slug":1350,"type":15},"Blazor","blazor",{"name":1352,"slug":1353,"type":15},"C#","csharp",{"name":1355,"slug":1356,"type":15},"Web Development","web-development",23787,"https:\u002F\u002Fgithub.com\u002Fopenai\u002Fskills","2026-04-12T05:07:02.819491",{"slug":1361,"name":1361,"fn":1362,"description":1363,"org":1364,"tags":1365,"stars":1357,"repoUrl":1358,"updatedAt":1376},"chatgpt-apps","build ChatGPT Apps SDK applications","Build, scaffold, refactor, and troubleshoot ChatGPT Apps SDK applications that combine an MCP server and widget UI. Use when Codex needs to design tools, register UI resources, wire the MCP Apps bridge or ChatGPT compatibility APIs, apply Apps SDK metadata or CSP or domain settings, or produce a docs-aligned project scaffold. Prefer a docs-first workflow by invoking the openai-docs skill or OpenAI developer docs MCP tools before generating code.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[1366,1369,1372,1375],{"name":1367,"slug":1368,"type":15},"Apps SDK","apps-sdk",{"name":1370,"slug":1371,"type":15},"ChatGPT","chatgpt",{"name":1373,"slug":1374,"type":15},"MCP","mcp",{"name":9,"slug":8,"type":15},"2026-04-12T05:07:05.468097",{"slug":1378,"name":1378,"fn":1379,"description":1380,"org":1381,"tags":1382,"stars":1357,"repoUrl":1358,"updatedAt":1392},"cli-creator","build CLIs from API docs","Build a composable CLI for Codex from API docs, an OpenAPI spec, existing curl examples, an SDK, a web app, an admin tool, or a local script. Use when the user wants Codex to create a command-line tool that can run from any repo, expose composable read\u002Fwrite commands, return stable JSON, manage auth, and pair with a companion skill.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[1383,1386,1389],{"name":1384,"slug":1385,"type":15},"API Development","api-development",{"name":1387,"slug":1388,"type":15},"CLI","cli",{"name":1390,"slug":1391,"type":15},"Codex","codex","2026-04-12T05:07:04.132762",{"slug":1394,"name":1394,"fn":1395,"description":1396,"org":1397,"tags":1398,"stars":1357,"repoUrl":1358,"updatedAt":1409},"cloudflare-deploy","deploy projects to Cloudflare","Deploy applications and infrastructure to Cloudflare using Workers, Pages, and related platform services. Use when the user asks to deploy, host, publish, or set up a project on Cloudflare.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[1399,1402,1405,1406],{"name":1400,"slug":1401,"type":15},"Cloudflare","cloudflare",{"name":1403,"slug":1404,"type":15},"Cloudflare Pages","cloudflare-pages",{"name":1247,"slug":1248,"type":15},{"name":1407,"slug":1408,"type":15},"Deployment","deployment","2026-04-12T05:07:14.275118",{"slug":1411,"name":1411,"fn":1412,"description":1413,"org":1414,"tags":1415,"stars":1357,"repoUrl":1358,"updatedAt":1428},"define-goal","define and set measurable project goals","Help the user define a concrete, measurable goal before starting work, especially when they ask to use the goal tool, create a goal, set an objective, clarify success criteria, or turn a fuzzy intention into a quantitative outcome. Use this skill for goal creation and goal refinement only; it does not manage durable snapshots, decision logs, or long-running execution artifacts.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[1416,1419,1422,1425],{"name":1417,"slug":1418,"type":15},"Productivity","productivity",{"name":1420,"slug":1421,"type":15},"Project Management","project-management",{"name":1423,"slug":1424,"type":15},"Strategy","strategy",{"name":1426,"slug":1427,"type":15},"Task Management","task-management","2026-05-23T06:17:16.870838",{"slug":1430,"name":1430,"fn":1431,"description":1432,"org":1433,"tags":1434,"stars":1357,"repoUrl":1358,"updatedAt":1440},"figma","translate Figma designs into code","Use the Figma MCP server to fetch design context, screenshots, variables, and assets from Figma, and to translate Figma nodes into production code. Trigger when a task involves Figma URLs, node IDs, design-to-code implementation, or Figma MCP setup and troubleshooting.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[1435,1436,1438,1439],{"name":1211,"slug":1212,"type":15},{"name":1437,"slug":1430,"type":15},"Figma",{"name":1266,"slug":1267,"type":15},{"name":1373,"slug":1374,"type":15},"2026-04-12T05:06:47.939943",{"slug":1442,"name":1442,"fn":1443,"description":1444,"org":1445,"tags":1446,"stars":1357,"repoUrl":1358,"updatedAt":1454},"figma-code-connect-components","connect Figma designs to code components","Connects Figma design components to code components using Code Connect mapping tools. Use when user says \"code connect\", \"connect this component to code\", \"map this component\", \"link component to code\", \"create code connect mapping\", or wants to establish mappings between Figma designs and code implementations. For canvas writes via `use_figma`, use `figma-use`.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[1447,1448,1451,1452,1453],{"name":1211,"slug":1212,"type":15},{"name":1449,"slug":1450,"type":15},"Design System","design-system",{"name":1437,"slug":1430,"type":15},{"name":1266,"slug":1267,"type":15},{"name":1275,"slug":1276,"type":15},"2026-05-10T05:59:52.971881",{"slug":1456,"name":1456,"fn":1457,"description":1458,"org":1459,"tags":1460,"stars":1357,"repoUrl":1358,"updatedAt":1468},"figma-create-design-system-rules","generate design system rules from Figma","Generates custom design system rules for the user's codebase. Use when user says \"create design system rules\", \"generate rules for my project\", \"set up design rules\", \"customize design system guidelines\", or wants to establish project-specific conventions for Figma-to-code workflows. Requires Figma MCP server connection.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[1461,1462,1463,1466,1467],{"name":1211,"slug":1212,"type":15},{"name":1449,"slug":1450,"type":15},{"name":1464,"slug":1465,"type":15},"Documentation","documentation",{"name":1437,"slug":1430,"type":15},{"name":1266,"slug":1267,"type":15},"2026-05-16T06:07:47.821474",{"slug":1470,"name":1470,"fn":1471,"description":1472,"org":1473,"tags":1474,"stars":1357,"repoUrl":1358,"updatedAt":1480},"figma-implement-design","translate Figma designs into application code","Translates Figma designs into production-ready application code with 1:1 visual fidelity. Use when implementing UI code from Figma files, when user mentions \"implement design\", \"generate code\", \"implement component\", provides Figma URLs, or asks to build components matching Figma specs. For Figma canvas writes via `use_figma`, use `figma-use`.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[1475,1476,1477,1478,1479],{"name":1211,"slug":1212,"type":15},{"name":1437,"slug":1430,"type":15},{"name":1266,"slug":1267,"type":15},{"name":1275,"slug":1276,"type":15},{"name":1355,"slug":1356,"type":15},"2026-05-16T06:07:40.583615",{"slug":1482,"name":1482,"fn":1483,"description":1484,"org":1485,"tags":1486,"stars":1357,"repoUrl":1358,"updatedAt":1495},"hatch-pet","create animated pets for Codex","Create, repair, validate, visually QA, and package Codex-compatible animated pets and pet spritesheets from character art, generated images, company or prospect brand cues, or visual references. Use when a user wants a lightweight-worker Codex pet workflow, a non-pixel custom pet style, a prospect or company mascot pet, or a full 8x9 animated pet atlas with transparent unused cells, QA contact sheets, and pet.json packaging. This skill composes the installed $imagegen system skill for visual generation and uses bundled scripts for deterministic spritesheet assembly.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[1487,1490,1491,1494],{"name":1488,"slug":1489,"type":15},"Animation","animation",{"name":1390,"slug":1391,"type":15},{"name":1492,"slug":1493,"type":15},"Creative","creative",{"name":1211,"slug":1212,"type":15},"2026-05-02T05:31:48.48485",{"slug":1497,"name":1497,"fn":1498,"description":1499,"org":1500,"tags":1501,"stars":1357,"repoUrl":1358,"updatedAt":1511},"imagegen","generate and edit raster images","Generate or edit raster images when the task benefits from AI-created bitmap visuals such as photos, illustrations, textures, sprites, mockups, or transparent-background cutouts. Use when Codex should create a brand-new image, transform an existing image, or derive visual variants from references, and the output should be a bitmap asset rather than repo-native code or vector. Do not use when the task is better handled by editing existing SVG\u002Fvector\u002Fcode-native assets, extending an established icon or logo system, or building the visual directly in HTML\u002FCSS\u002Fcanvas.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[1502,1503,1504,1507,1510],{"name":1492,"slug":1493,"type":15},{"name":1211,"slug":1212,"type":15},{"name":1505,"slug":1506,"type":15},"Image Generation","image-generation",{"name":1508,"slug":1509,"type":15},"Images","images",{"name":9,"slug":8,"type":15},"2026-05-15T06:23:24.312127",675]