[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill-vercel-labs-email-agent":3,"mdc-mwua2b-key":36,"related-org-vercel-labs-email-agent":2747,"related-repo-vercel-labs-email-agent":2919},{"slug":4,"name":4,"fn":5,"description":6,"org":7,"tags":11,"stars":25,"repoUrl":26,"updatedAt":27,"license":28,"forks":29,"topics":30,"repo":31,"sourceUrl":34,"mdContent":35},"email-agent","build AI-powered outbound email agents","Helps users understand and extend the Email Agent repo — an AI-powered outbound email generation app using Exa research, the Vercel AI SDK, and Outreach CRM. Use this skill when a user asks how the codebase works, how to hook up or swap an integration (Exa, OpenAI\u002FAI models, Outreach, Neon Postgres), or how to add a feature. Loaded automatically for any question about this repository.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},"vercel-labs","Vercel Labs","https:\u002F\u002Fpexgzepcugksgbtrxkhf.supabase.co\u002Fstorage\u002Fv1\u002Fobject\u002Fpublic\u002Forg-logos\u002Fvercel-labs.png",[12,16,19,22],{"name":13,"slug":14,"type":15},"Research","research","tag",{"name":17,"slug":18,"type":15},"CRM","crm",{"name":20,"slug":21,"type":15},"Agents","agents",{"name":23,"slug":24,"type":15},"Email","email",20,"https:\u002F\u002Fgithub.com\u002Fvercel-labs\u002Femail-agent","2026-07-17T06:04:19.880141",null,2,[],{"repoUrl":26,"stars":25,"forks":29,"topics":32,"description":33},[],"AI generated emails, set up directly in Outreach for SDRs","https:\u002F\u002Fgithub.com\u002Fvercel-labs\u002Femail-agent\u002Ftree\u002FHEAD\u002Fskills\u002Femail-agent","---\nname: email-agent\ndescription: Helps users understand and extend the Email Agent repo — an AI-powered outbound email generation app using Exa research, the Vercel AI SDK, and Outreach CRM. Use this skill when a user asks how the codebase works, how to hook up or swap an integration (Exa, OpenAI\u002FAI models, Outreach, Neon Postgres), or how to add a feature. Loaded automatically for any question about this repository.\n---\n\n# Email Agent Repo Guide\n\nThis skill gives you deep knowledge of the **Email Agent** codebase so you can answer \"how do I…\" questions, explain the architecture, and guide extensions — without re-reading every file each time.\n\n## What This Repo Is\n\nAn AI-powered outbound email generator. A user creates a **campaign** (with a custom AI prompt + research toggles), uploads **contacts** via CSV or API, and each contact is processed through a durable Vercel Workflow that:\n\n1. Researches the company (and optionally the person) via **Exa**\n2. Generates a personalized email sequence via the **Vercel AI SDK** (Anthropic models)\n3. Optionally syncs the prospect to **Outreach** CRM and enrolls them in a sequence\n\n**Stack:** Next.js 15 (App Router) · React Query · Vercel AI SDK · Drizzle ORM + Neon Postgres · Exa · Outreach · Tailwind 4 + shadcn\u002Fui · Vercel Workflows.\n\n## Architecture Map\n\n| Path | What lives here |\n|---|---|\n| `app\u002F` | Next.js pages + API routes. `\u002F` redirects to `\u002Fcampaigns`. `\u002Fsettings` has Outreach OAuth connect. |\n| `app\u002Fcampaigns\u002F` | Campaign list, new, and `[id]` detail pages (SSR + React Query hydration). |\n| `app\u002Fapi\u002Fcontacts\u002Froute.ts` | `POST \u002Fapi\u002Fcontacts` — external entry point to add a contact + auto-start its workflow. |\n| `app\u002Fapi\u002Foutreach\u002Fcallback\u002Froute.ts` | OAuth2 callback for Outreach; exchanges code for tokens, upserts into `oauth_tokens` table. |\n| `services\u002F` | Server actions (`'use server'`) for all internal DB operations. Called from client components via React Query. |\n| `workflows\u002Fprocess-contact.ts` | The durable workflow that orchestrates research → generation → Outreach for one contact. |\n| `lib\u002Fresearch\u002F` | Company + people research using Exa + AI summarization. |\n| `lib\u002Femail\u002F` | Email generation prompt builder + Zod output schema. |\n| `lib\u002Foutreach\u002F` | Outreach REST client, prospect upsert, sequence enrollment, sequence scaffolding. |\n| `lib\u002Fexa.ts` | Exa client singleton (reads `EXA_API_KEY`). |\n| `lib\u002Fcsv.ts` | CSV parsing with header aliasing and validation (Zod not used here; manual validation). |\n| `db\u002Fschema.ts` | Drizzle schema: `campaigns`, `contacts`, `oauth_tokens`, `settings` + TypeScript types. |\n| `db\u002Findex.ts` | Lazy Neon\u002FDrizzle proxy so the DB only initializes on first use. |\n| `components\u002F` | React components. shadcn\u002Fui primitives in `components\u002Fui\u002F`. |\n| `tests\u002F` | Vitest integration tests. `workflow.test.ts` hits real Exa + AI; only Outreach is mocked. |\n\n## The Core Workflow\n\n`workflows\u002Fprocess-contact.ts` is the heart of the app. Each contact runs through it independently and concurrently (`Promise.all` over `start(processContactWorkflow, [id])`).\n\n```\npending → researching → generating → sending → completed (or failed)\n```\n\nSteps (each is a `\"use step\"` function for durability + retries):\n1. **Load** contact + campaign from DB (`stepLoadContactAndCampaign`) — `FatalError` if missing\n2. **Research company** via Exa → summarized by AI (`stepResearchCompany`) — only if `campaign.researchEnabled`\n3. **Research person** via Exa → verified by AI (`stepResearchPerson`) — only if `campaign.peopleResearchEnabled`\n4. **Save research** to `contacts.companyResearch` \u002F `peopleResearch` (JSONB)\n5. **Generate email** via `generateText` with structured `Output.object` (`stepGenerateEmail`)\n6. **Save email** to `contacts.generatedSubject` + `generatedBody1\u002F2\u002F3`\n7. **Outreach** (if `campaign.outreachMode !== 'none'`): upsert prospect → set custom fields → optionally add to sequence\n8. **Mark completed** (or `failed` with error message in `contacts.data.error`)\n\nWorkflow launch points:\n- `services\u002Fcontacts.ts:processAllContacts` — \"Process N pending\" button\n- `services\u002Fcontacts.ts:addAndProcessContacts` — after CSV upload\n- `app\u002Fapi\u002Fcontacts\u002Froute.ts` — after API contact creation\n\n## External Integrations & APIs\n\n### Exa (research) — `lib\u002Fexa.ts`, `lib\u002Fresearch\u002F`\n\n- **Env var:** `EXA_API_KEY`\n- **Client:** `exa-js` SDK, singleton via `getExaClient()`\n- **Calls:** `exa.searchAndContents(query, { numResults, type: 'auto', category, summary: true })`\n- **Company research** (`lib\u002Fresearch\u002Fcompany.ts`): two parallel searches (`category: 'company'` + `category: 'news'`), results summarized by `anthropic\u002Fclaude-haiku-4-5` into `{ companySummary, existingAIFeatures[] }`\n- **People research** (`lib\u002Fresearch\u002Fpeople.ts`): two parallel searches (profile + activity), AI verifies `isConfidentMatch` (company+name must match) → returns `{ title, contactSummary, recentActivity[] }` or `null`\n- **Error handling:** both research functions catch errors and return safe fallbacks (empty research \u002F null) rather than failing the workflow\n\n### Vercel AI SDK (email generation + summarization) — `lib\u002Femail\u002F`\n\n- **Env var:** `AI_GATEWAY_API_KEY` (Vercel AI Gateway)\n- **Email generation** (`workflows\u002Fprocess-contact.ts:stepGenerateEmail`): `generateText({ model: 'anthropic\u002Fclaude-opus-4.6', output: Output.object({ schema }), prompt })`\n- **Research summarization:** `anthropic\u002Fclaude-haiku-4-5` with `Output.object` + Zod schema\n- **Prompt builder:** `lib\u002Femail\u002Fgeneration.ts:buildEmailGenerationPrompt` assembles sections: contact info, instructions (campaign system prompt), company research, people research, format requirements\n- **Output schema:** `lib\u002Femail\u002Fschema.ts:createEmailGenerationSchema(n)` — dynamically builds Zod object with `subject` + `body1..bodyN` (max 2 follow-ups = 3 bodies)\n- **Default prompt:** `DEFAULT_SYSTEM_PROMPT` in `lib\u002Femail\u002Fgeneration.ts` — 3-email F-shaped sequence, ~75 words each, HTML `\u003Cp>` formatting\n\n### Outreach (CRM) — `lib\u002Foutreach\u002F`\n\n- **Env vars:** `OUTREACH_CLIENT_ID`, `OUTREACH_CLIENT_SECRET`, `OUTREACH_REDIRECT_URI`\n- **OAuth flow:** `services\u002Foutreach.ts:getOutreachAuthUrl` → user authorizes at `api.outreach.io\u002Foauth\u002Fauthorize` → callback at `\u002Fapi\u002Foauth\u002Foutreach\u002Fcallback` → token exchange → stored in `oauth_tokens` table\n- **Token refresh:** `lib\u002Foutreach\u002Fclient.ts:refreshAccessToken` auto-refreshes if within 5 min of expiry\n- **API base:** `https:\u002F\u002Fapi.outreach.io`, JSON:API format (`Content-Type: application\u002Fvnd.api+json`)\n- **Custom fields:** subject → `custom92`, bodies → `custom93\u002F94\u002F95` (hardcoded in `lib\u002Foutreach\u002Fprospects.ts` and `sequences.ts`)\n- **Endpoints used:**\n  - `GET \u002Fapi\u002Fv2\u002Fprospects?filter[emails]=…` — find by email\n  - `POST \u002Fapi\u002Fv2\u002Fprospects` — create prospect\n  - `PATCH \u002Fapi\u002Fv2\u002Fprospects\u002F:id` — set custom fields\n  - `POST \u002Fapi\u002Fv2\u002FsequenceStates` — add prospect to sequence\n  - `POST \u002Fapi\u002Fv2\u002Fsequences`, `\u002FsequenceSteps`, `\u002Ftemplates`, `\u002FsequenceTemplates` — `scaffoldOutreachSequence` builds a full sequence\n\n### Neon Postgres — `db\u002F`\n\n- **Env var:** `DATABASE_URL`\n- **ORM:** Drizzle (`drizzle-orm\u002Fneon-http`), schema in `db\u002Fschema.ts`\n- **Migrations:** `pnpm db:push` (push) or `pnpm db:generate` (generate SQL)\n- **Tables:** `campaigns`, `contacts` (FK → campaigns, cascade delete), `oauth_tokens`, `settings`\n\n## How To Extend — Common Questions\n\n### \"How do I hook this up to a different CRM (instead of Outreach)?\"\n\nMirror the Outreach pattern:\n1. Create `lib\u002F\u003Ccrm>\u002Fclient.ts` — auth + authenticated `fetch` wrapper (see `lib\u002Foutreach\u002Fclient.ts`)\n2. Create `lib\u002F\u003Ccrm>\u002Fprospects.ts` — `upsertProspect`, `setProspectCustomFields`, `addProspectToSequence`\n3. Add an OAuth callback route at `app\u002Fapi\u002F\u003Ccrm>\u002Fcallback\u002Froute.ts` (see the Outreach callback)\n4. Add a server action in `services\u002F` for generating the auth URL + checking connection\n5. Add a `\u003Ccrm>Mode` enum to `db\u002Fschema.ts` alongside the existing `outreachMode` enum, then branch in `workflows\u002Fprocess-contact.ts` (the workflow already checks `campaign.outreachMode`)\n6. Add a settings UI component for the OAuth connect button\n\nThe workflow is the integration point — `stepUpsertProspect`, `stepSetProspectFields`, `stepAddToSequence` are isolated steps you can swap.\n\n### \"How do I use a different research provider (instead of Exa)?\"\n\n1. Replace `lib\u002Fexa.ts` with a new client, or create `lib\u002F\u003Cprovider>.ts`\n2. Rewrite `lib\u002Fresearch\u002Fcompany.ts` and `lib\u002Fresearch\u002Fpeople.ts` to call your provider's search API\n3. Keep the return types `CompanyResearch` and `PeopleResearch` (defined in `db\u002Fschema.ts`) so the workflow and prompt builder don't need changes\n4. Update the env var from `EXA_API_KEY` to your provider's key\n\nThe research functions are self-contained — the workflow just calls `researchCompany()` \u002F `researchPerson()` and saves the result.\n\n### \"How do I use a different AI model?\"\n\nThe model is a string passed to `generateText({ model: '…' })`. The app uses the Vercel AI Gateway, so any model the gateway supports works:\n- Email generation: `anthropic\u002Fclaude-opus-4.6` in `workflows\u002Fprocess-contact.ts:222`\n- Research summarization: `anthropic\u002Fclaude-haiku-4-5` in `lib\u002Fresearch\u002Fcompany.ts:53` and `lib\u002Fresearch\u002Fpeople.ts:70`\n\nJust change the model string. The `Output.object({ schema })` structured output pattern works across providers.\n\n### \"How do I add contacts from another source (webhook, CRM sync, etc.)?\"\n\nTwo options:\n- **API route:** `POST \u002Fapi\u002Fcontacts` already exists (`app\u002Fapi\u002Fcontacts\u002Froute.ts`). It takes `{ campaignId, contact: { email, firstName, lastName?, companyName, context? } }` and auto-starts the workflow. Point any external system at this endpoint.\n- **Server action:** Call `services\u002Fcontacts.ts:addAndProcessContacts(campaignId, contactsData)` directly from a server component or another server action.\n\n### \"How do I add more follow-up emails?\"\n\nCurrently capped at 2 follow-ups (3 total) by `MAX_FOLLOW_UPS` in `lib\u002Femail\u002Fschema.ts`. To increase:\n1. Bump `MAX_FOLLOW_UPS`\n2. Add `generatedBody4`, etc. columns to `contacts` in `db\u002Fschema.ts` → run `pnpm db:push`\n3. Add more Outreach custom field slots in `lib\u002Foutreach\u002Fprospects.ts` (`CUSTOM_FIELD_BODIES`) and `sequences.ts`\n4. Update `stepSaveEmail` in the workflow to save the extra bodies\n5. Update the campaign form select options in `components\u002Fcampaign-form.tsx`\n\n### \"How do I add a new campaign setting?\"\n\n1. Add the column to `campaigns` in `db\u002Fschema.ts` → `pnpm db:push`\n2. Add the form field in `components\u002Fcampaign-form.tsx`\n3. Read it in `workflows\u002Fprocess-contact.ts` (the campaign object is loaded in `stepLoadContactAndCampaign`)\n4. If it affects the prompt, pass it through `buildEmailGenerationPrompt` in `lib\u002Femail\u002Fgeneration.ts`\n\n## Coding Conventions (quick reference — full detail in README.md)\n\n- **Server actions** (`services\u002F`, `'use server'`) for internal data ops. **API routes** (`app\u002Fapi\u002F`) only for external entry points (OAuth, webhooks, third-party callers).\n- **Vercel Workflows** for anything calling external APIs or doing AI generation. Use `\"use workflow\"` + `\"use step\"`. `FatalError` = non-retryable; regular `throw` = retryable.\n- **React Query** for all client data. Query keys centralized in `lib\u002Fquery-keys.ts`. Poll at 3s when contacts are processing. Pass SSR data as `initialData`.\n- **Drizzle** types via `InferSelectModel` \u002F `$inferInsert` — don't duplicate. JSONB for complex nested data. `onDelete: 'cascade'` for children. `withTimezone: true` timestamps.\n- **Components:** server components by default, `'use client'` only when needed. shadcn\u002Fui for primitives. Native `FormData` for forms (no form libraries).\n- **TypeScript:** strict mode, prefer inference, validate untrusted data with Zod, no `any`.\n- **Tests:** integration tests in `tests\u002F` with Vitest, `fileParallelism: false`, real Exa + AI calls (only Outreach mocked). Use `createTracker()` for cleanup.\n\n## Commands\n\n```bash\npnpm dev          # local dev server\npnpm build        # production build\npnpm lint         # next lint\npnpm test         # vitest run (needs .env.local with real API keys + DATABASE_URL)\npnpm test:watch   # vitest watch mode\npnpm db:push      # push schema changes to Neon\npnpm db:generate  # generate migration SQL\npnpm db:studio    # Drizzle Studio GUI\npnpm workflows    # inspect workflow runs (npx workflow inspect runs --web)\n```\n\n## Environment Variables\n\n| Var | Purpose |\n|---|---|\n| `AI_GATEWAY_API_KEY` | Vercel AI Gateway — required for all AI calls |\n| `EXA_API_KEY` | Exa search API — required for research |\n| `DATABASE_URL` | Neon Postgres connection string |\n| `OUTREACH_CLIENT_ID` | Outreach OAuth (optional — only if using Outreach) |\n| `OUTREACH_CLIENT_SECRET` | Outreach OAuth (optional) |\n| `OUTREACH_REDIRECT_URI` | Must be `https:\u002F\u002F\u003Cdomain>\u002Fapi\u002Foauth\u002Foutreach\u002Fcallback` |\n| `NEXT_PUBLIC_SITE_URL` | Used for the API curl example on campaign detail page |\n\n## Answering Questions\n\nWhen a user asks \"how do I…\" about this repo:\n1. Identify which integration\u002Flayer they're asking about (research, AI, CRM, DB, UI, workflow).\n2. Reference the exact files above so they can navigate directly.\n3. If they want to swap an integration, point them to the \"How To Extend\" section that matches and name the specific files to change.\n4. If unsure about current behavior, read the specific file before answering — this guide is a map, not a substitute for the code.\n",{"data":37,"body":38},{"name":4,"description":6},{"type":39,"children":40},"root",[41,50,64,71,90,129,139,145,535,541,566,578,591,792,797,833,839,857,1008,1019,1166,1177,1426,1438,1536,1542,1548,1553,1687,1714,1720,1798,1818,1824,1837,1882,1895,1901,1906,1958,1964,1984,2076,2082,2153,2159,2378,2384,2560,2566,2707,2713,2718,2741],{"type":42,"tag":43,"props":44,"children":46},"element","h1",{"id":45},"email-agent-repo-guide",[47],{"type":48,"value":49},"text","Email Agent Repo Guide",{"type":42,"tag":51,"props":52,"children":53},"p",{},[54,56,62],{"type":48,"value":55},"This skill gives you deep knowledge of the ",{"type":42,"tag":57,"props":58,"children":59},"strong",{},[60],{"type":48,"value":61},"Email Agent",{"type":48,"value":63}," codebase so you can answer \"how do I…\" questions, explain the architecture, and guide extensions — without re-reading every file each time.",{"type":42,"tag":65,"props":66,"children":68},"h2",{"id":67},"what-this-repo-is",[69],{"type":48,"value":70},"What This Repo Is",{"type":42,"tag":51,"props":72,"children":73},{},[74,76,81,83,88],{"type":48,"value":75},"An AI-powered outbound email generator. A user creates a ",{"type":42,"tag":57,"props":77,"children":78},{},[79],{"type":48,"value":80},"campaign",{"type":48,"value":82}," (with a custom AI prompt + research toggles), uploads ",{"type":42,"tag":57,"props":84,"children":85},{},[86],{"type":48,"value":87},"contacts",{"type":48,"value":89}," via CSV or API, and each contact is processed through a durable Vercel Workflow that:",{"type":42,"tag":91,"props":92,"children":93},"ol",{},[94,105,117],{"type":42,"tag":95,"props":96,"children":97},"li",{},[98,100],{"type":48,"value":99},"Researches the company (and optionally the person) via ",{"type":42,"tag":57,"props":101,"children":102},{},[103],{"type":48,"value":104},"Exa",{"type":42,"tag":95,"props":106,"children":107},{},[108,110,115],{"type":48,"value":109},"Generates a personalized email sequence via the ",{"type":42,"tag":57,"props":111,"children":112},{},[113],{"type":48,"value":114},"Vercel AI SDK",{"type":48,"value":116}," (Anthropic models)",{"type":42,"tag":95,"props":118,"children":119},{},[120,122,127],{"type":48,"value":121},"Optionally syncs the prospect to ",{"type":42,"tag":57,"props":123,"children":124},{},[125],{"type":48,"value":126},"Outreach",{"type":48,"value":128}," CRM and enrolls them in a sequence",{"type":42,"tag":51,"props":130,"children":131},{},[132,137],{"type":42,"tag":57,"props":133,"children":134},{},[135],{"type":48,"value":136},"Stack:",{"type":48,"value":138}," Next.js 15 (App Router) · React Query · Vercel AI SDK · Drizzle ORM + Neon Postgres · Exa · Outreach · Tailwind 4 + shadcn\u002Fui · Vercel Workflows.",{"type":42,"tag":65,"props":140,"children":142},{"id":141},"architecture-map",[143],{"type":48,"value":144},"Architecture Map",{"type":42,"tag":146,"props":147,"children":148},"table",{},[149,168],{"type":42,"tag":150,"props":151,"children":152},"thead",{},[153],{"type":42,"tag":154,"props":155,"children":156},"tr",{},[157,163],{"type":42,"tag":158,"props":159,"children":160},"th",{},[161],{"type":48,"value":162},"Path",{"type":42,"tag":158,"props":164,"children":165},{},[166],{"type":48,"value":167},"What lives here",{"type":42,"tag":169,"props":170,"children":171},"tbody",{},[172,215,240,263,288,313,330,347,364,381,406,423,468,485,510],{"type":42,"tag":154,"props":173,"children":174},{},[175,186],{"type":42,"tag":176,"props":177,"children":178},"td",{},[179],{"type":42,"tag":180,"props":181,"children":183},"code",{"className":182},[],[184],{"type":48,"value":185},"app\u002F",{"type":42,"tag":176,"props":187,"children":188},{},[189,191,197,199,205,207,213],{"type":48,"value":190},"Next.js pages + API routes. ",{"type":42,"tag":180,"props":192,"children":194},{"className":193},[],[195],{"type":48,"value":196},"\u002F",{"type":48,"value":198}," redirects to ",{"type":42,"tag":180,"props":200,"children":202},{"className":201},[],[203],{"type":48,"value":204},"\u002Fcampaigns",{"type":48,"value":206},". ",{"type":42,"tag":180,"props":208,"children":210},{"className":209},[],[211],{"type":48,"value":212},"\u002Fsettings",{"type":48,"value":214}," has Outreach OAuth connect.",{"type":42,"tag":154,"props":216,"children":217},{},[218,227],{"type":42,"tag":176,"props":219,"children":220},{},[221],{"type":42,"tag":180,"props":222,"children":224},{"className":223},[],[225],{"type":48,"value":226},"app\u002Fcampaigns\u002F",{"type":42,"tag":176,"props":228,"children":229},{},[230,232,238],{"type":48,"value":231},"Campaign list, new, and ",{"type":42,"tag":180,"props":233,"children":235},{"className":234},[],[236],{"type":48,"value":237},"[id]",{"type":48,"value":239}," detail pages (SSR + React Query hydration).",{"type":42,"tag":154,"props":241,"children":242},{},[243,252],{"type":42,"tag":176,"props":244,"children":245},{},[246],{"type":42,"tag":180,"props":247,"children":249},{"className":248},[],[250],{"type":48,"value":251},"app\u002Fapi\u002Fcontacts\u002Froute.ts",{"type":42,"tag":176,"props":253,"children":254},{},[255,261],{"type":42,"tag":180,"props":256,"children":258},{"className":257},[],[259],{"type":48,"value":260},"POST \u002Fapi\u002Fcontacts",{"type":48,"value":262}," — external entry point to add a contact + auto-start its workflow.",{"type":42,"tag":154,"props":264,"children":265},{},[266,275],{"type":42,"tag":176,"props":267,"children":268},{},[269],{"type":42,"tag":180,"props":270,"children":272},{"className":271},[],[273],{"type":48,"value":274},"app\u002Fapi\u002Foutreach\u002Fcallback\u002Froute.ts",{"type":42,"tag":176,"props":276,"children":277},{},[278,280,286],{"type":48,"value":279},"OAuth2 callback for Outreach; exchanges code for tokens, upserts into ",{"type":42,"tag":180,"props":281,"children":283},{"className":282},[],[284],{"type":48,"value":285},"oauth_tokens",{"type":48,"value":287}," table.",{"type":42,"tag":154,"props":289,"children":290},{},[291,300],{"type":42,"tag":176,"props":292,"children":293},{},[294],{"type":42,"tag":180,"props":295,"children":297},{"className":296},[],[298],{"type":48,"value":299},"services\u002F",{"type":42,"tag":176,"props":301,"children":302},{},[303,305,311],{"type":48,"value":304},"Server actions (",{"type":42,"tag":180,"props":306,"children":308},{"className":307},[],[309],{"type":48,"value":310},"'use server'",{"type":48,"value":312},") for all internal DB operations. Called from client components via React Query.",{"type":42,"tag":154,"props":314,"children":315},{},[316,325],{"type":42,"tag":176,"props":317,"children":318},{},[319],{"type":42,"tag":180,"props":320,"children":322},{"className":321},[],[323],{"type":48,"value":324},"workflows\u002Fprocess-contact.ts",{"type":42,"tag":176,"props":326,"children":327},{},[328],{"type":48,"value":329},"The durable workflow that orchestrates research → generation → Outreach for one contact.",{"type":42,"tag":154,"props":331,"children":332},{},[333,342],{"type":42,"tag":176,"props":334,"children":335},{},[336],{"type":42,"tag":180,"props":337,"children":339},{"className":338},[],[340],{"type":48,"value":341},"lib\u002Fresearch\u002F",{"type":42,"tag":176,"props":343,"children":344},{},[345],{"type":48,"value":346},"Company + people research using Exa + AI summarization.",{"type":42,"tag":154,"props":348,"children":349},{},[350,359],{"type":42,"tag":176,"props":351,"children":352},{},[353],{"type":42,"tag":180,"props":354,"children":356},{"className":355},[],[357],{"type":48,"value":358},"lib\u002Femail\u002F",{"type":42,"tag":176,"props":360,"children":361},{},[362],{"type":48,"value":363},"Email generation prompt builder + Zod output schema.",{"type":42,"tag":154,"props":365,"children":366},{},[367,376],{"type":42,"tag":176,"props":368,"children":369},{},[370],{"type":42,"tag":180,"props":371,"children":373},{"className":372},[],[374],{"type":48,"value":375},"lib\u002Foutreach\u002F",{"type":42,"tag":176,"props":377,"children":378},{},[379],{"type":48,"value":380},"Outreach REST client, prospect upsert, sequence enrollment, sequence scaffolding.",{"type":42,"tag":154,"props":382,"children":383},{},[384,393],{"type":42,"tag":176,"props":385,"children":386},{},[387],{"type":42,"tag":180,"props":388,"children":390},{"className":389},[],[391],{"type":48,"value":392},"lib\u002Fexa.ts",{"type":42,"tag":176,"props":394,"children":395},{},[396,398,404],{"type":48,"value":397},"Exa client singleton (reads ",{"type":42,"tag":180,"props":399,"children":401},{"className":400},[],[402],{"type":48,"value":403},"EXA_API_KEY",{"type":48,"value":405},").",{"type":42,"tag":154,"props":407,"children":408},{},[409,418],{"type":42,"tag":176,"props":410,"children":411},{},[412],{"type":42,"tag":180,"props":413,"children":415},{"className":414},[],[416],{"type":48,"value":417},"lib\u002Fcsv.ts",{"type":42,"tag":176,"props":419,"children":420},{},[421],{"type":48,"value":422},"CSV parsing with header aliasing and validation (Zod not used here; manual validation).",{"type":42,"tag":154,"props":424,"children":425},{},[426,435],{"type":42,"tag":176,"props":427,"children":428},{},[429],{"type":42,"tag":180,"props":430,"children":432},{"className":431},[],[433],{"type":48,"value":434},"db\u002Fschema.ts",{"type":42,"tag":176,"props":436,"children":437},{},[438,440,446,448,453,454,459,460,466],{"type":48,"value":439},"Drizzle schema: ",{"type":42,"tag":180,"props":441,"children":443},{"className":442},[],[444],{"type":48,"value":445},"campaigns",{"type":48,"value":447},", ",{"type":42,"tag":180,"props":449,"children":451},{"className":450},[],[452],{"type":48,"value":87},{"type":48,"value":447},{"type":42,"tag":180,"props":455,"children":457},{"className":456},[],[458],{"type":48,"value":285},{"type":48,"value":447},{"type":42,"tag":180,"props":461,"children":463},{"className":462},[],[464],{"type":48,"value":465},"settings",{"type":48,"value":467}," + TypeScript types.",{"type":42,"tag":154,"props":469,"children":470},{},[471,480],{"type":42,"tag":176,"props":472,"children":473},{},[474],{"type":42,"tag":180,"props":475,"children":477},{"className":476},[],[478],{"type":48,"value":479},"db\u002Findex.ts",{"type":42,"tag":176,"props":481,"children":482},{},[483],{"type":48,"value":484},"Lazy Neon\u002FDrizzle proxy so the DB only initializes on first use.",{"type":42,"tag":154,"props":486,"children":487},{},[488,497],{"type":42,"tag":176,"props":489,"children":490},{},[491],{"type":42,"tag":180,"props":492,"children":494},{"className":493},[],[495],{"type":48,"value":496},"components\u002F",{"type":42,"tag":176,"props":498,"children":499},{},[500,502,508],{"type":48,"value":501},"React components. shadcn\u002Fui primitives in ",{"type":42,"tag":180,"props":503,"children":505},{"className":504},[],[506],{"type":48,"value":507},"components\u002Fui\u002F",{"type":48,"value":509},".",{"type":42,"tag":154,"props":511,"children":512},{},[513,522],{"type":42,"tag":176,"props":514,"children":515},{},[516],{"type":42,"tag":180,"props":517,"children":519},{"className":518},[],[520],{"type":48,"value":521},"tests\u002F",{"type":42,"tag":176,"props":523,"children":524},{},[525,527,533],{"type":48,"value":526},"Vitest integration tests. ",{"type":42,"tag":180,"props":528,"children":530},{"className":529},[],[531],{"type":48,"value":532},"workflow.test.ts",{"type":48,"value":534}," hits real Exa + AI; only Outreach is mocked.",{"type":42,"tag":65,"props":536,"children":538},{"id":537},"the-core-workflow",[539],{"type":48,"value":540},"The Core Workflow",{"type":42,"tag":51,"props":542,"children":543},{},[544,549,551,557,559,565],{"type":42,"tag":180,"props":545,"children":547},{"className":546},[],[548],{"type":48,"value":324},{"type":48,"value":550}," is the heart of the app. Each contact runs through it independently and concurrently (",{"type":42,"tag":180,"props":552,"children":554},{"className":553},[],[555],{"type":48,"value":556},"Promise.all",{"type":48,"value":558}," over ",{"type":42,"tag":180,"props":560,"children":562},{"className":561},[],[563],{"type":48,"value":564},"start(processContactWorkflow, [id])",{"type":48,"value":405},{"type":42,"tag":567,"props":568,"children":572},"pre",{"className":569,"code":571,"language":48},[570],"language-text","pending → researching → generating → sending → completed (or failed)\n",[573],{"type":42,"tag":180,"props":574,"children":576},{"__ignoreMap":575},"",[577],{"type":48,"value":571},{"type":42,"tag":51,"props":579,"children":580},{},[581,583,589],{"type":48,"value":582},"Steps (each is a ",{"type":42,"tag":180,"props":584,"children":586},{"className":585},[],[587],{"type":48,"value":588},"\"use step\"",{"type":48,"value":590}," function for durability + retries):",{"type":42,"tag":91,"props":592,"children":593},{},[594,620,644,667,693,727,750,767],{"type":42,"tag":95,"props":595,"children":596},{},[597,602,604,610,612,618],{"type":42,"tag":57,"props":598,"children":599},{},[600],{"type":48,"value":601},"Load",{"type":48,"value":603}," contact + campaign from DB (",{"type":42,"tag":180,"props":605,"children":607},{"className":606},[],[608],{"type":48,"value":609},"stepLoadContactAndCampaign",{"type":48,"value":611},") — ",{"type":42,"tag":180,"props":613,"children":615},{"className":614},[],[616],{"type":48,"value":617},"FatalError",{"type":48,"value":619}," if missing",{"type":42,"tag":95,"props":621,"children":622},{},[623,628,630,636,638],{"type":42,"tag":57,"props":624,"children":625},{},[626],{"type":48,"value":627},"Research company",{"type":48,"value":629}," via Exa → summarized by AI (",{"type":42,"tag":180,"props":631,"children":633},{"className":632},[],[634],{"type":48,"value":635},"stepResearchCompany",{"type":48,"value":637},") — only if ",{"type":42,"tag":180,"props":639,"children":641},{"className":640},[],[642],{"type":48,"value":643},"campaign.researchEnabled",{"type":42,"tag":95,"props":645,"children":646},{},[647,652,654,660,661],{"type":42,"tag":57,"props":648,"children":649},{},[650],{"type":48,"value":651},"Research person",{"type":48,"value":653}," via Exa → verified by AI (",{"type":42,"tag":180,"props":655,"children":657},{"className":656},[],[658],{"type":48,"value":659},"stepResearchPerson",{"type":48,"value":637},{"type":42,"tag":180,"props":662,"children":664},{"className":663},[],[665],{"type":48,"value":666},"campaign.peopleResearchEnabled",{"type":42,"tag":95,"props":668,"children":669},{},[670,675,677,683,685,691],{"type":42,"tag":57,"props":671,"children":672},{},[673],{"type":48,"value":674},"Save research",{"type":48,"value":676}," to ",{"type":42,"tag":180,"props":678,"children":680},{"className":679},[],[681],{"type":48,"value":682},"contacts.companyResearch",{"type":48,"value":684}," \u002F ",{"type":42,"tag":180,"props":686,"children":688},{"className":687},[],[689],{"type":48,"value":690},"peopleResearch",{"type":48,"value":692}," (JSONB)",{"type":42,"tag":95,"props":694,"children":695},{},[696,701,703,709,711,717,719,725],{"type":42,"tag":57,"props":697,"children":698},{},[699],{"type":48,"value":700},"Generate email",{"type":48,"value":702}," via ",{"type":42,"tag":180,"props":704,"children":706},{"className":705},[],[707],{"type":48,"value":708},"generateText",{"type":48,"value":710}," with structured ",{"type":42,"tag":180,"props":712,"children":714},{"className":713},[],[715],{"type":48,"value":716},"Output.object",{"type":48,"value":718}," (",{"type":42,"tag":180,"props":720,"children":722},{"className":721},[],[723],{"type":48,"value":724},"stepGenerateEmail",{"type":48,"value":726},")",{"type":42,"tag":95,"props":728,"children":729},{},[730,735,736,742,744],{"type":42,"tag":57,"props":731,"children":732},{},[733],{"type":48,"value":734},"Save email",{"type":48,"value":676},{"type":42,"tag":180,"props":737,"children":739},{"className":738},[],[740],{"type":48,"value":741},"contacts.generatedSubject",{"type":48,"value":743}," + ",{"type":42,"tag":180,"props":745,"children":747},{"className":746},[],[748],{"type":48,"value":749},"generatedBody1\u002F2\u002F3",{"type":42,"tag":95,"props":751,"children":752},{},[753,757,759,765],{"type":42,"tag":57,"props":754,"children":755},{},[756],{"type":48,"value":126},{"type":48,"value":758}," (if ",{"type":42,"tag":180,"props":760,"children":762},{"className":761},[],[763],{"type":48,"value":764},"campaign.outreachMode !== 'none'",{"type":48,"value":766},"): upsert prospect → set custom fields → optionally add to sequence",{"type":42,"tag":95,"props":768,"children":769},{},[770,775,777,783,785,791],{"type":42,"tag":57,"props":771,"children":772},{},[773],{"type":48,"value":774},"Mark completed",{"type":48,"value":776}," (or ",{"type":42,"tag":180,"props":778,"children":780},{"className":779},[],[781],{"type":48,"value":782},"failed",{"type":48,"value":784}," with error message in ",{"type":42,"tag":180,"props":786,"children":788},{"className":787},[],[789],{"type":48,"value":790},"contacts.data.error",{"type":48,"value":726},{"type":42,"tag":51,"props":793,"children":794},{},[795],{"type":48,"value":796},"Workflow launch points:",{"type":42,"tag":798,"props":799,"children":800},"ul",{},[801,812,823],{"type":42,"tag":95,"props":802,"children":803},{},[804,810],{"type":42,"tag":180,"props":805,"children":807},{"className":806},[],[808],{"type":48,"value":809},"services\u002Fcontacts.ts:processAllContacts",{"type":48,"value":811}," — \"Process N pending\" button",{"type":42,"tag":95,"props":813,"children":814},{},[815,821],{"type":42,"tag":180,"props":816,"children":818},{"className":817},[],[819],{"type":48,"value":820},"services\u002Fcontacts.ts:addAndProcessContacts",{"type":48,"value":822}," — after CSV upload",{"type":42,"tag":95,"props":824,"children":825},{},[826,831],{"type":42,"tag":180,"props":827,"children":829},{"className":828},[],[830],{"type":48,"value":251},{"type":48,"value":832}," — after API contact creation",{"type":42,"tag":65,"props":834,"children":836},{"id":835},"external-integrations-apis",[837],{"type":48,"value":838},"External Integrations & APIs",{"type":42,"tag":840,"props":841,"children":843},"h3",{"id":842},"exa-research-libexats-libresearch",[844,846,851,852],{"type":48,"value":845},"Exa (research) — ",{"type":42,"tag":180,"props":847,"children":849},{"className":848},[],[850],{"type":48,"value":392},{"type":48,"value":447},{"type":42,"tag":180,"props":853,"children":855},{"className":854},[],[856],{"type":48,"value":341},{"type":42,"tag":798,"props":858,"children":859},{},[860,875,898,913,959,998],{"type":42,"tag":95,"props":861,"children":862},{},[863,868,870],{"type":42,"tag":57,"props":864,"children":865},{},[866],{"type":48,"value":867},"Env var:",{"type":48,"value":869}," ",{"type":42,"tag":180,"props":871,"children":873},{"className":872},[],[874],{"type":48,"value":403},{"type":42,"tag":95,"props":876,"children":877},{},[878,883,884,890,892],{"type":42,"tag":57,"props":879,"children":880},{},[881],{"type":48,"value":882},"Client:",{"type":48,"value":869},{"type":42,"tag":180,"props":885,"children":887},{"className":886},[],[888],{"type":48,"value":889},"exa-js",{"type":48,"value":891}," SDK, singleton via ",{"type":42,"tag":180,"props":893,"children":895},{"className":894},[],[896],{"type":48,"value":897},"getExaClient()",{"type":42,"tag":95,"props":899,"children":900},{},[901,906,907],{"type":42,"tag":57,"props":902,"children":903},{},[904],{"type":48,"value":905},"Calls:",{"type":48,"value":869},{"type":42,"tag":180,"props":908,"children":910},{"className":909},[],[911],{"type":48,"value":912},"exa.searchAndContents(query, { numResults, type: 'auto', category, summary: true })",{"type":42,"tag":95,"props":914,"children":915},{},[916,921,922,928,930,936,937,943,945,951,953],{"type":42,"tag":57,"props":917,"children":918},{},[919],{"type":48,"value":920},"Company research",{"type":48,"value":718},{"type":42,"tag":180,"props":923,"children":925},{"className":924},[],[926],{"type":48,"value":927},"lib\u002Fresearch\u002Fcompany.ts",{"type":48,"value":929},"): two parallel searches (",{"type":42,"tag":180,"props":931,"children":933},{"className":932},[],[934],{"type":48,"value":935},"category: 'company'",{"type":48,"value":743},{"type":42,"tag":180,"props":938,"children":940},{"className":939},[],[941],{"type":48,"value":942},"category: 'news'",{"type":48,"value":944},"), results summarized by ",{"type":42,"tag":180,"props":946,"children":948},{"className":947},[],[949],{"type":48,"value":950},"anthropic\u002Fclaude-haiku-4-5",{"type":48,"value":952}," into ",{"type":42,"tag":180,"props":954,"children":956},{"className":955},[],[957],{"type":48,"value":958},"{ companySummary, existingAIFeatures[] }",{"type":42,"tag":95,"props":960,"children":961},{},[962,967,968,974,976,982,984,990,992],{"type":42,"tag":57,"props":963,"children":964},{},[965],{"type":48,"value":966},"People research",{"type":48,"value":718},{"type":42,"tag":180,"props":969,"children":971},{"className":970},[],[972],{"type":48,"value":973},"lib\u002Fresearch\u002Fpeople.ts",{"type":48,"value":975},"): two parallel searches (profile + activity), AI verifies ",{"type":42,"tag":180,"props":977,"children":979},{"className":978},[],[980],{"type":48,"value":981},"isConfidentMatch",{"type":48,"value":983}," (company+name must match) → returns ",{"type":42,"tag":180,"props":985,"children":987},{"className":986},[],[988],{"type":48,"value":989},"{ title, contactSummary, recentActivity[] }",{"type":48,"value":991}," or ",{"type":42,"tag":180,"props":993,"children":995},{"className":994},[],[996],{"type":48,"value":997},"null",{"type":42,"tag":95,"props":999,"children":1000},{},[1001,1006],{"type":42,"tag":57,"props":1002,"children":1003},{},[1004],{"type":48,"value":1005},"Error handling:",{"type":48,"value":1007}," both research functions catch errors and return safe fallbacks (empty research \u002F null) rather than failing the workflow",{"type":42,"tag":840,"props":1009,"children":1011},{"id":1010},"vercel-ai-sdk-email-generation-summarization-libemail",[1012,1014],{"type":48,"value":1013},"Vercel AI SDK (email generation + summarization) — ",{"type":42,"tag":180,"props":1015,"children":1017},{"className":1016},[],[1018],{"type":48,"value":358},{"type":42,"tag":798,"props":1020,"children":1021},{},[1022,1038,1061,1084,1101,1133],{"type":42,"tag":95,"props":1023,"children":1024},{},[1025,1029,1030,1036],{"type":42,"tag":57,"props":1026,"children":1027},{},[1028],{"type":48,"value":867},{"type":48,"value":869},{"type":42,"tag":180,"props":1031,"children":1033},{"className":1032},[],[1034],{"type":48,"value":1035},"AI_GATEWAY_API_KEY",{"type":48,"value":1037}," (Vercel AI Gateway)",{"type":42,"tag":95,"props":1039,"children":1040},{},[1041,1046,1047,1053,1055],{"type":42,"tag":57,"props":1042,"children":1043},{},[1044],{"type":48,"value":1045},"Email generation",{"type":48,"value":718},{"type":42,"tag":180,"props":1048,"children":1050},{"className":1049},[],[1051],{"type":48,"value":1052},"workflows\u002Fprocess-contact.ts:stepGenerateEmail",{"type":48,"value":1054},"): ",{"type":42,"tag":180,"props":1056,"children":1058},{"className":1057},[],[1059],{"type":48,"value":1060},"generateText({ model: 'anthropic\u002Fclaude-opus-4.6', output: Output.object({ schema }), prompt })",{"type":42,"tag":95,"props":1062,"children":1063},{},[1064,1069,1070,1075,1077,1082],{"type":42,"tag":57,"props":1065,"children":1066},{},[1067],{"type":48,"value":1068},"Research summarization:",{"type":48,"value":869},{"type":42,"tag":180,"props":1071,"children":1073},{"className":1072},[],[1074],{"type":48,"value":950},{"type":48,"value":1076}," with ",{"type":42,"tag":180,"props":1078,"children":1080},{"className":1079},[],[1081],{"type":48,"value":716},{"type":48,"value":1083}," + Zod schema",{"type":42,"tag":95,"props":1085,"children":1086},{},[1087,1092,1093,1099],{"type":42,"tag":57,"props":1088,"children":1089},{},[1090],{"type":48,"value":1091},"Prompt builder:",{"type":48,"value":869},{"type":42,"tag":180,"props":1094,"children":1096},{"className":1095},[],[1097],{"type":48,"value":1098},"lib\u002Femail\u002Fgeneration.ts:buildEmailGenerationPrompt",{"type":48,"value":1100}," assembles sections: contact info, instructions (campaign system prompt), company research, people research, format requirements",{"type":42,"tag":95,"props":1102,"children":1103},{},[1104,1109,1110,1116,1118,1124,1125,1131],{"type":42,"tag":57,"props":1105,"children":1106},{},[1107],{"type":48,"value":1108},"Output schema:",{"type":48,"value":869},{"type":42,"tag":180,"props":1111,"children":1113},{"className":1112},[],[1114],{"type":48,"value":1115},"lib\u002Femail\u002Fschema.ts:createEmailGenerationSchema(n)",{"type":48,"value":1117}," — dynamically builds Zod object with ",{"type":42,"tag":180,"props":1119,"children":1121},{"className":1120},[],[1122],{"type":48,"value":1123},"subject",{"type":48,"value":743},{"type":42,"tag":180,"props":1126,"children":1128},{"className":1127},[],[1129],{"type":48,"value":1130},"body1..bodyN",{"type":48,"value":1132}," (max 2 follow-ups = 3 bodies)",{"type":42,"tag":95,"props":1134,"children":1135},{},[1136,1141,1142,1148,1150,1156,1158,1164],{"type":42,"tag":57,"props":1137,"children":1138},{},[1139],{"type":48,"value":1140},"Default prompt:",{"type":48,"value":869},{"type":42,"tag":180,"props":1143,"children":1145},{"className":1144},[],[1146],{"type":48,"value":1147},"DEFAULT_SYSTEM_PROMPT",{"type":48,"value":1149}," in ",{"type":42,"tag":180,"props":1151,"children":1153},{"className":1152},[],[1154],{"type":48,"value":1155},"lib\u002Femail\u002Fgeneration.ts",{"type":48,"value":1157}," — 3-email F-shaped sequence, ~75 words each, HTML ",{"type":42,"tag":180,"props":1159,"children":1161},{"className":1160},[],[1162],{"type":48,"value":1163},"\u003Cp>",{"type":48,"value":1165}," formatting",{"type":42,"tag":840,"props":1167,"children":1169},{"id":1168},"outreach-crm-liboutreach",[1170,1172],{"type":48,"value":1171},"Outreach (CRM) — ",{"type":42,"tag":180,"props":1173,"children":1175},{"className":1174},[],[1176],{"type":48,"value":375},{"type":42,"tag":798,"props":1178,"children":1179},{},[1180,1209,1249,1266,1290,1331],{"type":42,"tag":95,"props":1181,"children":1182},{},[1183,1188,1189,1195,1196,1202,1203],{"type":42,"tag":57,"props":1184,"children":1185},{},[1186],{"type":48,"value":1187},"Env vars:",{"type":48,"value":869},{"type":42,"tag":180,"props":1190,"children":1192},{"className":1191},[],[1193],{"type":48,"value":1194},"OUTREACH_CLIENT_ID",{"type":48,"value":447},{"type":42,"tag":180,"props":1197,"children":1199},{"className":1198},[],[1200],{"type":48,"value":1201},"OUTREACH_CLIENT_SECRET",{"type":48,"value":447},{"type":42,"tag":180,"props":1204,"children":1206},{"className":1205},[],[1207],{"type":48,"value":1208},"OUTREACH_REDIRECT_URI",{"type":42,"tag":95,"props":1210,"children":1211},{},[1212,1217,1218,1224,1226,1232,1234,1240,1242,1247],{"type":42,"tag":57,"props":1213,"children":1214},{},[1215],{"type":48,"value":1216},"OAuth flow:",{"type":48,"value":869},{"type":42,"tag":180,"props":1219,"children":1221},{"className":1220},[],[1222],{"type":48,"value":1223},"services\u002Foutreach.ts:getOutreachAuthUrl",{"type":48,"value":1225}," → user authorizes at ",{"type":42,"tag":180,"props":1227,"children":1229},{"className":1228},[],[1230],{"type":48,"value":1231},"api.outreach.io\u002Foauth\u002Fauthorize",{"type":48,"value":1233}," → callback at ",{"type":42,"tag":180,"props":1235,"children":1237},{"className":1236},[],[1238],{"type":48,"value":1239},"\u002Fapi\u002Foauth\u002Foutreach\u002Fcallback",{"type":48,"value":1241}," → token exchange → stored in ",{"type":42,"tag":180,"props":1243,"children":1245},{"className":1244},[],[1246],{"type":48,"value":285},{"type":48,"value":1248}," table",{"type":42,"tag":95,"props":1250,"children":1251},{},[1252,1257,1258,1264],{"type":42,"tag":57,"props":1253,"children":1254},{},[1255],{"type":48,"value":1256},"Token refresh:",{"type":48,"value":869},{"type":42,"tag":180,"props":1259,"children":1261},{"className":1260},[],[1262],{"type":48,"value":1263},"lib\u002Foutreach\u002Fclient.ts:refreshAccessToken",{"type":48,"value":1265}," auto-refreshes if within 5 min of expiry",{"type":42,"tag":95,"props":1267,"children":1268},{},[1269,1274,1275,1281,1283,1289],{"type":42,"tag":57,"props":1270,"children":1271},{},[1272],{"type":48,"value":1273},"API base:",{"type":48,"value":869},{"type":42,"tag":180,"props":1276,"children":1278},{"className":1277},[],[1279],{"type":48,"value":1280},"https:\u002F\u002Fapi.outreach.io",{"type":48,"value":1282},", JSON:API format (",{"type":42,"tag":180,"props":1284,"children":1286},{"className":1285},[],[1287],{"type":48,"value":1288},"Content-Type: application\u002Fvnd.api+json",{"type":48,"value":726},{"type":42,"tag":95,"props":1291,"children":1292},{},[1293,1298,1300,1306,1308,1314,1316,1322,1324,1330],{"type":42,"tag":57,"props":1294,"children":1295},{},[1296],{"type":48,"value":1297},"Custom fields:",{"type":48,"value":1299}," subject → ",{"type":42,"tag":180,"props":1301,"children":1303},{"className":1302},[],[1304],{"type":48,"value":1305},"custom92",{"type":48,"value":1307},", bodies → ",{"type":42,"tag":180,"props":1309,"children":1311},{"className":1310},[],[1312],{"type":48,"value":1313},"custom93\u002F94\u002F95",{"type":48,"value":1315}," (hardcoded in ",{"type":42,"tag":180,"props":1317,"children":1319},{"className":1318},[],[1320],{"type":48,"value":1321},"lib\u002Foutreach\u002Fprospects.ts",{"type":48,"value":1323}," and ",{"type":42,"tag":180,"props":1325,"children":1327},{"className":1326},[],[1328],{"type":48,"value":1329},"sequences.ts",{"type":48,"value":726},{"type":42,"tag":95,"props":1332,"children":1333},{},[1334,1339],{"type":42,"tag":57,"props":1335,"children":1336},{},[1337],{"type":48,"value":1338},"Endpoints used:",{"type":42,"tag":798,"props":1340,"children":1341},{},[1342,1353,1364,1375,1386],{"type":42,"tag":95,"props":1343,"children":1344},{},[1345,1351],{"type":42,"tag":180,"props":1346,"children":1348},{"className":1347},[],[1349],{"type":48,"value":1350},"GET \u002Fapi\u002Fv2\u002Fprospects?filter[emails]=…",{"type":48,"value":1352}," — find by email",{"type":42,"tag":95,"props":1354,"children":1355},{},[1356,1362],{"type":42,"tag":180,"props":1357,"children":1359},{"className":1358},[],[1360],{"type":48,"value":1361},"POST \u002Fapi\u002Fv2\u002Fprospects",{"type":48,"value":1363}," — create prospect",{"type":42,"tag":95,"props":1365,"children":1366},{},[1367,1373],{"type":42,"tag":180,"props":1368,"children":1370},{"className":1369},[],[1371],{"type":48,"value":1372},"PATCH \u002Fapi\u002Fv2\u002Fprospects\u002F:id",{"type":48,"value":1374}," — set custom fields",{"type":42,"tag":95,"props":1376,"children":1377},{},[1378,1384],{"type":42,"tag":180,"props":1379,"children":1381},{"className":1380},[],[1382],{"type":48,"value":1383},"POST \u002Fapi\u002Fv2\u002FsequenceStates",{"type":48,"value":1385}," — add prospect to sequence",{"type":42,"tag":95,"props":1387,"children":1388},{},[1389,1395,1396,1402,1403,1409,1410,1416,1418,1424],{"type":42,"tag":180,"props":1390,"children":1392},{"className":1391},[],[1393],{"type":48,"value":1394},"POST \u002Fapi\u002Fv2\u002Fsequences",{"type":48,"value":447},{"type":42,"tag":180,"props":1397,"children":1399},{"className":1398},[],[1400],{"type":48,"value":1401},"\u002FsequenceSteps",{"type":48,"value":447},{"type":42,"tag":180,"props":1404,"children":1406},{"className":1405},[],[1407],{"type":48,"value":1408},"\u002Ftemplates",{"type":48,"value":447},{"type":42,"tag":180,"props":1411,"children":1413},{"className":1412},[],[1414],{"type":48,"value":1415},"\u002FsequenceTemplates",{"type":48,"value":1417}," — ",{"type":42,"tag":180,"props":1419,"children":1421},{"className":1420},[],[1422],{"type":48,"value":1423},"scaffoldOutreachSequence",{"type":48,"value":1425}," builds a full sequence",{"type":42,"tag":840,"props":1427,"children":1429},{"id":1428},"neon-postgres-db",[1430,1432],{"type":48,"value":1431},"Neon Postgres — ",{"type":42,"tag":180,"props":1433,"children":1435},{"className":1434},[],[1436],{"type":48,"value":1437},"db\u002F",{"type":42,"tag":798,"props":1439,"children":1440},{},[1441,1455,1478,1503],{"type":42,"tag":95,"props":1442,"children":1443},{},[1444,1448,1449],{"type":42,"tag":57,"props":1445,"children":1446},{},[1447],{"type":48,"value":867},{"type":48,"value":869},{"type":42,"tag":180,"props":1450,"children":1452},{"className":1451},[],[1453],{"type":48,"value":1454},"DATABASE_URL",{"type":42,"tag":95,"props":1456,"children":1457},{},[1458,1463,1465,1471,1473],{"type":42,"tag":57,"props":1459,"children":1460},{},[1461],{"type":48,"value":1462},"ORM:",{"type":48,"value":1464}," Drizzle (",{"type":42,"tag":180,"props":1466,"children":1468},{"className":1467},[],[1469],{"type":48,"value":1470},"drizzle-orm\u002Fneon-http",{"type":48,"value":1472},"), schema in ",{"type":42,"tag":180,"props":1474,"children":1476},{"className":1475},[],[1477],{"type":48,"value":434},{"type":42,"tag":95,"props":1479,"children":1480},{},[1481,1486,1487,1493,1495,1501],{"type":42,"tag":57,"props":1482,"children":1483},{},[1484],{"type":48,"value":1485},"Migrations:",{"type":48,"value":869},{"type":42,"tag":180,"props":1488,"children":1490},{"className":1489},[],[1491],{"type":48,"value":1492},"pnpm db:push",{"type":48,"value":1494}," (push) or ",{"type":42,"tag":180,"props":1496,"children":1498},{"className":1497},[],[1499],{"type":48,"value":1500},"pnpm db:generate",{"type":48,"value":1502}," (generate SQL)",{"type":42,"tag":95,"props":1504,"children":1505},{},[1506,1511,1512,1517,1518,1523,1525,1530,1531],{"type":42,"tag":57,"props":1507,"children":1508},{},[1509],{"type":48,"value":1510},"Tables:",{"type":48,"value":869},{"type":42,"tag":180,"props":1513,"children":1515},{"className":1514},[],[1516],{"type":48,"value":445},{"type":48,"value":447},{"type":42,"tag":180,"props":1519,"children":1521},{"className":1520},[],[1522],{"type":48,"value":87},{"type":48,"value":1524}," (FK → campaigns, cascade delete), ",{"type":42,"tag":180,"props":1526,"children":1528},{"className":1527},[],[1529],{"type":48,"value":285},{"type":48,"value":447},{"type":42,"tag":180,"props":1532,"children":1534},{"className":1533},[],[1535],{"type":48,"value":465},{"type":42,"tag":65,"props":1537,"children":1539},{"id":1538},"how-to-extend-common-questions",[1540],{"type":48,"value":1541},"How To Extend — Common Questions",{"type":42,"tag":840,"props":1543,"children":1545},{"id":1544},"how-do-i-hook-this-up-to-a-different-crm-instead-of-outreach",[1546],{"type":48,"value":1547},"\"How do I hook this up to a different CRM (instead of Outreach)?\"",{"type":42,"tag":51,"props":1549,"children":1550},{},[1551],{"type":48,"value":1552},"Mirror the Outreach pattern:",{"type":42,"tag":91,"props":1554,"children":1555},{},[1556,1584,1615,1628,1640,1682],{"type":42,"tag":95,"props":1557,"children":1558},{},[1559,1561,1567,1569,1575,1577,1583],{"type":48,"value":1560},"Create ",{"type":42,"tag":180,"props":1562,"children":1564},{"className":1563},[],[1565],{"type":48,"value":1566},"lib\u002F\u003Ccrm>\u002Fclient.ts",{"type":48,"value":1568}," — auth + authenticated ",{"type":42,"tag":180,"props":1570,"children":1572},{"className":1571},[],[1573],{"type":48,"value":1574},"fetch",{"type":48,"value":1576}," wrapper (see ",{"type":42,"tag":180,"props":1578,"children":1580},{"className":1579},[],[1581],{"type":48,"value":1582},"lib\u002Foutreach\u002Fclient.ts",{"type":48,"value":726},{"type":42,"tag":95,"props":1585,"children":1586},{},[1587,1588,1594,1595,1601,1602,1608,1609],{"type":48,"value":1560},{"type":42,"tag":180,"props":1589,"children":1591},{"className":1590},[],[1592],{"type":48,"value":1593},"lib\u002F\u003Ccrm>\u002Fprospects.ts",{"type":48,"value":1417},{"type":42,"tag":180,"props":1596,"children":1598},{"className":1597},[],[1599],{"type":48,"value":1600},"upsertProspect",{"type":48,"value":447},{"type":42,"tag":180,"props":1603,"children":1605},{"className":1604},[],[1606],{"type":48,"value":1607},"setProspectCustomFields",{"type":48,"value":447},{"type":42,"tag":180,"props":1610,"children":1612},{"className":1611},[],[1613],{"type":48,"value":1614},"addProspectToSequence",{"type":42,"tag":95,"props":1616,"children":1617},{},[1618,1620,1626],{"type":48,"value":1619},"Add an OAuth callback route at ",{"type":42,"tag":180,"props":1621,"children":1623},{"className":1622},[],[1624],{"type":48,"value":1625},"app\u002Fapi\u002F\u003Ccrm>\u002Fcallback\u002Froute.ts",{"type":48,"value":1627}," (see the Outreach callback)",{"type":42,"tag":95,"props":1629,"children":1630},{},[1631,1633,1638],{"type":48,"value":1632},"Add a server action in ",{"type":42,"tag":180,"props":1634,"children":1636},{"className":1635},[],[1637],{"type":48,"value":299},{"type":48,"value":1639}," for generating the auth URL + checking connection",{"type":42,"tag":95,"props":1641,"children":1642},{},[1643,1645,1651,1653,1658,1660,1666,1668,1673,1675,1681],{"type":48,"value":1644},"Add a ",{"type":42,"tag":180,"props":1646,"children":1648},{"className":1647},[],[1649],{"type":48,"value":1650},"\u003Ccrm>Mode",{"type":48,"value":1652}," enum to ",{"type":42,"tag":180,"props":1654,"children":1656},{"className":1655},[],[1657],{"type":48,"value":434},{"type":48,"value":1659}," alongside the existing ",{"type":42,"tag":180,"props":1661,"children":1663},{"className":1662},[],[1664],{"type":48,"value":1665},"outreachMode",{"type":48,"value":1667}," enum, then branch in ",{"type":42,"tag":180,"props":1669,"children":1671},{"className":1670},[],[1672],{"type":48,"value":324},{"type":48,"value":1674}," (the workflow already checks ",{"type":42,"tag":180,"props":1676,"children":1678},{"className":1677},[],[1679],{"type":48,"value":1680},"campaign.outreachMode",{"type":48,"value":726},{"type":42,"tag":95,"props":1683,"children":1684},{},[1685],{"type":48,"value":1686},"Add a settings UI component for the OAuth connect button",{"type":42,"tag":51,"props":1688,"children":1689},{},[1690,1692,1698,1699,1705,1706,1712],{"type":48,"value":1691},"The workflow is the integration point — ",{"type":42,"tag":180,"props":1693,"children":1695},{"className":1694},[],[1696],{"type":48,"value":1697},"stepUpsertProspect",{"type":48,"value":447},{"type":42,"tag":180,"props":1700,"children":1702},{"className":1701},[],[1703],{"type":48,"value":1704},"stepSetProspectFields",{"type":48,"value":447},{"type":42,"tag":180,"props":1707,"children":1709},{"className":1708},[],[1710],{"type":48,"value":1711},"stepAddToSequence",{"type":48,"value":1713}," are isolated steps you can swap.",{"type":42,"tag":840,"props":1715,"children":1717},{"id":1716},"how-do-i-use-a-different-research-provider-instead-of-exa",[1718],{"type":48,"value":1719},"\"How do I use a different research provider (instead of Exa)?\"",{"type":42,"tag":91,"props":1721,"children":1722},{},[1723,1741,1759,1786],{"type":42,"tag":95,"props":1724,"children":1725},{},[1726,1728,1733,1735],{"type":48,"value":1727},"Replace ",{"type":42,"tag":180,"props":1729,"children":1731},{"className":1730},[],[1732],{"type":48,"value":392},{"type":48,"value":1734}," with a new client, or create ",{"type":42,"tag":180,"props":1736,"children":1738},{"className":1737},[],[1739],{"type":48,"value":1740},"lib\u002F\u003Cprovider>.ts",{"type":42,"tag":95,"props":1742,"children":1743},{},[1744,1746,1751,1752,1757],{"type":48,"value":1745},"Rewrite ",{"type":42,"tag":180,"props":1747,"children":1749},{"className":1748},[],[1750],{"type":48,"value":927},{"type":48,"value":1323},{"type":42,"tag":180,"props":1753,"children":1755},{"className":1754},[],[1756],{"type":48,"value":973},{"type":48,"value":1758}," to call your provider's search API",{"type":42,"tag":95,"props":1760,"children":1761},{},[1762,1764,1770,1771,1777,1779,1784],{"type":48,"value":1763},"Keep the return types ",{"type":42,"tag":180,"props":1765,"children":1767},{"className":1766},[],[1768],{"type":48,"value":1769},"CompanyResearch",{"type":48,"value":1323},{"type":42,"tag":180,"props":1772,"children":1774},{"className":1773},[],[1775],{"type":48,"value":1776},"PeopleResearch",{"type":48,"value":1778}," (defined in ",{"type":42,"tag":180,"props":1780,"children":1782},{"className":1781},[],[1783],{"type":48,"value":434},{"type":48,"value":1785},") so the workflow and prompt builder don't need changes",{"type":42,"tag":95,"props":1787,"children":1788},{},[1789,1791,1796],{"type":48,"value":1790},"Update the env var from ",{"type":42,"tag":180,"props":1792,"children":1794},{"className":1793},[],[1795],{"type":48,"value":403},{"type":48,"value":1797}," to your provider's key",{"type":42,"tag":51,"props":1799,"children":1800},{},[1801,1803,1809,1810,1816],{"type":48,"value":1802},"The research functions are self-contained — the workflow just calls ",{"type":42,"tag":180,"props":1804,"children":1806},{"className":1805},[],[1807],{"type":48,"value":1808},"researchCompany()",{"type":48,"value":684},{"type":42,"tag":180,"props":1811,"children":1813},{"className":1812},[],[1814],{"type":48,"value":1815},"researchPerson()",{"type":48,"value":1817}," and saves the result.",{"type":42,"tag":840,"props":1819,"children":1821},{"id":1820},"how-do-i-use-a-different-ai-model",[1822],{"type":48,"value":1823},"\"How do I use a different AI model?\"",{"type":42,"tag":51,"props":1825,"children":1826},{},[1827,1829,1835],{"type":48,"value":1828},"The model is a string passed to ",{"type":42,"tag":180,"props":1830,"children":1832},{"className":1831},[],[1833],{"type":48,"value":1834},"generateText({ model: '…' })",{"type":48,"value":1836},". The app uses the Vercel AI Gateway, so any model the gateway supports works:",{"type":42,"tag":798,"props":1838,"children":1839},{},[1840,1858],{"type":42,"tag":95,"props":1841,"children":1842},{},[1843,1845,1851,1852],{"type":48,"value":1844},"Email generation: ",{"type":42,"tag":180,"props":1846,"children":1848},{"className":1847},[],[1849],{"type":48,"value":1850},"anthropic\u002Fclaude-opus-4.6",{"type":48,"value":1149},{"type":42,"tag":180,"props":1853,"children":1855},{"className":1854},[],[1856],{"type":48,"value":1857},"workflows\u002Fprocess-contact.ts:222",{"type":42,"tag":95,"props":1859,"children":1860},{},[1861,1863,1868,1869,1875,1876],{"type":48,"value":1862},"Research summarization: ",{"type":42,"tag":180,"props":1864,"children":1866},{"className":1865},[],[1867],{"type":48,"value":950},{"type":48,"value":1149},{"type":42,"tag":180,"props":1870,"children":1872},{"className":1871},[],[1873],{"type":48,"value":1874},"lib\u002Fresearch\u002Fcompany.ts:53",{"type":48,"value":1323},{"type":42,"tag":180,"props":1877,"children":1879},{"className":1878},[],[1880],{"type":48,"value":1881},"lib\u002Fresearch\u002Fpeople.ts:70",{"type":42,"tag":51,"props":1883,"children":1884},{},[1885,1887,1893],{"type":48,"value":1886},"Just change the model string. The ",{"type":42,"tag":180,"props":1888,"children":1890},{"className":1889},[],[1891],{"type":48,"value":1892},"Output.object({ schema })",{"type":48,"value":1894}," structured output pattern works across providers.",{"type":42,"tag":840,"props":1896,"children":1898},{"id":1897},"how-do-i-add-contacts-from-another-source-webhook-crm-sync-etc",[1899],{"type":48,"value":1900},"\"How do I add contacts from another source (webhook, CRM sync, etc.)?\"",{"type":42,"tag":51,"props":1902,"children":1903},{},[1904],{"type":48,"value":1905},"Two options:",{"type":42,"tag":798,"props":1907,"children":1908},{},[1909,1940],{"type":42,"tag":95,"props":1910,"children":1911},{},[1912,1917,1918,1923,1925,1930,1932,1938],{"type":42,"tag":57,"props":1913,"children":1914},{},[1915],{"type":48,"value":1916},"API route:",{"type":48,"value":869},{"type":42,"tag":180,"props":1919,"children":1921},{"className":1920},[],[1922],{"type":48,"value":260},{"type":48,"value":1924}," already exists (",{"type":42,"tag":180,"props":1926,"children":1928},{"className":1927},[],[1929],{"type":48,"value":251},{"type":48,"value":1931},"). It takes ",{"type":42,"tag":180,"props":1933,"children":1935},{"className":1934},[],[1936],{"type":48,"value":1937},"{ campaignId, contact: { email, firstName, lastName?, companyName, context? } }",{"type":48,"value":1939}," and auto-starts the workflow. Point any external system at this endpoint.",{"type":42,"tag":95,"props":1941,"children":1942},{},[1943,1948,1950,1956],{"type":42,"tag":57,"props":1944,"children":1945},{},[1946],{"type":48,"value":1947},"Server action:",{"type":48,"value":1949}," Call ",{"type":42,"tag":180,"props":1951,"children":1953},{"className":1952},[],[1954],{"type":48,"value":1955},"services\u002Fcontacts.ts:addAndProcessContacts(campaignId, contactsData)",{"type":48,"value":1957}," directly from a server component or another server action.",{"type":42,"tag":840,"props":1959,"children":1961},{"id":1960},"how-do-i-add-more-follow-up-emails",[1962],{"type":48,"value":1963},"\"How do I add more follow-up emails?\"",{"type":42,"tag":51,"props":1965,"children":1966},{},[1967,1969,1975,1976,1982],{"type":48,"value":1968},"Currently capped at 2 follow-ups (3 total) by ",{"type":42,"tag":180,"props":1970,"children":1972},{"className":1971},[],[1973],{"type":48,"value":1974},"MAX_FOLLOW_UPS",{"type":48,"value":1149},{"type":42,"tag":180,"props":1977,"children":1979},{"className":1978},[],[1980],{"type":48,"value":1981},"lib\u002Femail\u002Fschema.ts",{"type":48,"value":1983},". To increase:",{"type":42,"tag":91,"props":1985,"children":1986},{},[1987,1997,2028,2052,2065],{"type":42,"tag":95,"props":1988,"children":1989},{},[1990,1992],{"type":48,"value":1991},"Bump ",{"type":42,"tag":180,"props":1993,"children":1995},{"className":1994},[],[1996],{"type":48,"value":1974},{"type":42,"tag":95,"props":1998,"children":1999},{},[2000,2002,2008,2010,2015,2016,2021,2023],{"type":48,"value":2001},"Add ",{"type":42,"tag":180,"props":2003,"children":2005},{"className":2004},[],[2006],{"type":48,"value":2007},"generatedBody4",{"type":48,"value":2009},", etc. columns to ",{"type":42,"tag":180,"props":2011,"children":2013},{"className":2012},[],[2014],{"type":48,"value":87},{"type":48,"value":1149},{"type":42,"tag":180,"props":2017,"children":2019},{"className":2018},[],[2020],{"type":48,"value":434},{"type":48,"value":2022}," → run ",{"type":42,"tag":180,"props":2024,"children":2026},{"className":2025},[],[2027],{"type":48,"value":1492},{"type":42,"tag":95,"props":2029,"children":2030},{},[2031,2033,2038,2039,2045,2047],{"type":48,"value":2032},"Add more Outreach custom field slots in ",{"type":42,"tag":180,"props":2034,"children":2036},{"className":2035},[],[2037],{"type":48,"value":1321},{"type":48,"value":718},{"type":42,"tag":180,"props":2040,"children":2042},{"className":2041},[],[2043],{"type":48,"value":2044},"CUSTOM_FIELD_BODIES",{"type":48,"value":2046},") and ",{"type":42,"tag":180,"props":2048,"children":2050},{"className":2049},[],[2051],{"type":48,"value":1329},{"type":42,"tag":95,"props":2053,"children":2054},{},[2055,2057,2063],{"type":48,"value":2056},"Update ",{"type":42,"tag":180,"props":2058,"children":2060},{"className":2059},[],[2061],{"type":48,"value":2062},"stepSaveEmail",{"type":48,"value":2064}," in the workflow to save the extra bodies",{"type":42,"tag":95,"props":2066,"children":2067},{},[2068,2070],{"type":48,"value":2069},"Update the campaign form select options in ",{"type":42,"tag":180,"props":2071,"children":2073},{"className":2072},[],[2074],{"type":48,"value":2075},"components\u002Fcampaign-form.tsx",{"type":42,"tag":840,"props":2077,"children":2079},{"id":2078},"how-do-i-add-a-new-campaign-setting",[2080],{"type":48,"value":2081},"\"How do I add a new campaign setting?\"",{"type":42,"tag":91,"props":2083,"children":2084},{},[2085,2108,2118,2136],{"type":42,"tag":95,"props":2086,"children":2087},{},[2088,2090,2095,2096,2101,2103],{"type":48,"value":2089},"Add the column to ",{"type":42,"tag":180,"props":2091,"children":2093},{"className":2092},[],[2094],{"type":48,"value":445},{"type":48,"value":1149},{"type":42,"tag":180,"props":2097,"children":2099},{"className":2098},[],[2100],{"type":48,"value":434},{"type":48,"value":2102}," → ",{"type":42,"tag":180,"props":2104,"children":2106},{"className":2105},[],[2107],{"type":48,"value":1492},{"type":42,"tag":95,"props":2109,"children":2110},{},[2111,2113],{"type":48,"value":2112},"Add the form field in ",{"type":42,"tag":180,"props":2114,"children":2116},{"className":2115},[],[2117],{"type":48,"value":2075},{"type":42,"tag":95,"props":2119,"children":2120},{},[2121,2123,2128,2130,2135],{"type":48,"value":2122},"Read it in ",{"type":42,"tag":180,"props":2124,"children":2126},{"className":2125},[],[2127],{"type":48,"value":324},{"type":48,"value":2129}," (the campaign object is loaded in ",{"type":42,"tag":180,"props":2131,"children":2133},{"className":2132},[],[2134],{"type":48,"value":609},{"type":48,"value":726},{"type":42,"tag":95,"props":2137,"children":2138},{},[2139,2141,2147,2148],{"type":48,"value":2140},"If it affects the prompt, pass it through ",{"type":42,"tag":180,"props":2142,"children":2144},{"className":2143},[],[2145],{"type":48,"value":2146},"buildEmailGenerationPrompt",{"type":48,"value":1149},{"type":42,"tag":180,"props":2149,"children":2151},{"className":2150},[],[2152],{"type":48,"value":1155},{"type":42,"tag":65,"props":2154,"children":2156},{"id":2155},"coding-conventions-quick-reference-full-detail-in-readmemd",[2157],{"type":48,"value":2158},"Coding Conventions (quick reference — full detail in README.md)",{"type":42,"tag":798,"props":2160,"children":2161},{},[2162,2198,2236,2261,2302,2328,2345],{"type":42,"tag":95,"props":2163,"children":2164},{},[2165,2170,2171,2176,2177,2182,2184,2189,2190,2196],{"type":42,"tag":57,"props":2166,"children":2167},{},[2168],{"type":48,"value":2169},"Server actions",{"type":48,"value":718},{"type":42,"tag":180,"props":2172,"children":2174},{"className":2173},[],[2175],{"type":48,"value":299},{"type":48,"value":447},{"type":42,"tag":180,"props":2178,"children":2180},{"className":2179},[],[2181],{"type":48,"value":310},{"type":48,"value":2183},") for internal data ops. ",{"type":42,"tag":57,"props":2185,"children":2186},{},[2187],{"type":48,"value":2188},"API routes",{"type":48,"value":718},{"type":42,"tag":180,"props":2191,"children":2193},{"className":2192},[],[2194],{"type":48,"value":2195},"app\u002Fapi\u002F",{"type":48,"value":2197},") only for external entry points (OAuth, webhooks, third-party callers).",{"type":42,"tag":95,"props":2199,"children":2200},{},[2201,2206,2208,2214,2215,2220,2221,2226,2228,2234],{"type":42,"tag":57,"props":2202,"children":2203},{},[2204],{"type":48,"value":2205},"Vercel Workflows",{"type":48,"value":2207}," for anything calling external APIs or doing AI generation. Use ",{"type":42,"tag":180,"props":2209,"children":2211},{"className":2210},[],[2212],{"type":48,"value":2213},"\"use workflow\"",{"type":48,"value":743},{"type":42,"tag":180,"props":2216,"children":2218},{"className":2217},[],[2219],{"type":48,"value":588},{"type":48,"value":206},{"type":42,"tag":180,"props":2222,"children":2224},{"className":2223},[],[2225],{"type":48,"value":617},{"type":48,"value":2227}," = non-retryable; regular ",{"type":42,"tag":180,"props":2229,"children":2231},{"className":2230},[],[2232],{"type":48,"value":2233},"throw",{"type":48,"value":2235}," = retryable.",{"type":42,"tag":95,"props":2237,"children":2238},{},[2239,2244,2246,2252,2254,2260],{"type":42,"tag":57,"props":2240,"children":2241},{},[2242],{"type":48,"value":2243},"React Query",{"type":48,"value":2245}," for all client data. Query keys centralized in ",{"type":42,"tag":180,"props":2247,"children":2249},{"className":2248},[],[2250],{"type":48,"value":2251},"lib\u002Fquery-keys.ts",{"type":48,"value":2253},". Poll at 3s when contacts are processing. Pass SSR data as ",{"type":42,"tag":180,"props":2255,"children":2257},{"className":2256},[],[2258],{"type":48,"value":2259},"initialData",{"type":48,"value":509},{"type":42,"tag":95,"props":2262,"children":2263},{},[2264,2269,2271,2277,2278,2284,2286,2292,2294,2300],{"type":42,"tag":57,"props":2265,"children":2266},{},[2267],{"type":48,"value":2268},"Drizzle",{"type":48,"value":2270}," types via ",{"type":42,"tag":180,"props":2272,"children":2274},{"className":2273},[],[2275],{"type":48,"value":2276},"InferSelectModel",{"type":48,"value":684},{"type":42,"tag":180,"props":2279,"children":2281},{"className":2280},[],[2282],{"type":48,"value":2283},"$inferInsert",{"type":48,"value":2285}," — don't duplicate. JSONB for complex nested data. ",{"type":42,"tag":180,"props":2287,"children":2289},{"className":2288},[],[2290],{"type":48,"value":2291},"onDelete: 'cascade'",{"type":48,"value":2293}," for children. ",{"type":42,"tag":180,"props":2295,"children":2297},{"className":2296},[],[2298],{"type":48,"value":2299},"withTimezone: true",{"type":48,"value":2301}," timestamps.",{"type":42,"tag":95,"props":2303,"children":2304},{},[2305,2310,2312,2318,2320,2326],{"type":42,"tag":57,"props":2306,"children":2307},{},[2308],{"type":48,"value":2309},"Components:",{"type":48,"value":2311}," server components by default, ",{"type":42,"tag":180,"props":2313,"children":2315},{"className":2314},[],[2316],{"type":48,"value":2317},"'use client'",{"type":48,"value":2319}," only when needed. shadcn\u002Fui for primitives. Native ",{"type":42,"tag":180,"props":2321,"children":2323},{"className":2322},[],[2324],{"type":48,"value":2325},"FormData",{"type":48,"value":2327}," for forms (no form libraries).",{"type":42,"tag":95,"props":2329,"children":2330},{},[2331,2336,2338,2344],{"type":42,"tag":57,"props":2332,"children":2333},{},[2334],{"type":48,"value":2335},"TypeScript:",{"type":48,"value":2337}," strict mode, prefer inference, validate untrusted data with Zod, no ",{"type":42,"tag":180,"props":2339,"children":2341},{"className":2340},[],[2342],{"type":48,"value":2343},"any",{"type":48,"value":509},{"type":42,"tag":95,"props":2346,"children":2347},{},[2348,2353,2355,2360,2362,2368,2370,2376],{"type":42,"tag":57,"props":2349,"children":2350},{},[2351],{"type":48,"value":2352},"Tests:",{"type":48,"value":2354}," integration tests in ",{"type":42,"tag":180,"props":2356,"children":2358},{"className":2357},[],[2359],{"type":48,"value":521},{"type":48,"value":2361}," with Vitest, ",{"type":42,"tag":180,"props":2363,"children":2365},{"className":2364},[],[2366],{"type":48,"value":2367},"fileParallelism: false",{"type":48,"value":2369},", real Exa + AI calls (only Outreach mocked). Use ",{"type":42,"tag":180,"props":2371,"children":2373},{"className":2372},[],[2374],{"type":48,"value":2375},"createTracker()",{"type":48,"value":2377}," for cleanup.",{"type":42,"tag":65,"props":2379,"children":2381},{"id":2380},"commands",[2382],{"type":48,"value":2383},"Commands",{"type":42,"tag":567,"props":2385,"children":2389},{"className":2386,"code":2387,"language":2388,"meta":575,"style":575},"language-bash shiki shiki-themes material-theme-lighter material-theme material-theme-palenight","pnpm dev          # local dev server\npnpm build        # production build\npnpm lint         # next lint\npnpm test         # vitest run (needs .env.local with real API keys + DATABASE_URL)\npnpm test:watch   # vitest watch mode\npnpm db:push      # push schema changes to Neon\npnpm db:generate  # generate migration SQL\npnpm db:studio    # Drizzle Studio GUI\npnpm workflows    # inspect workflow runs (npx workflow inspect runs --web)\n","bash",[2390],{"type":42,"tag":180,"props":2391,"children":2392},{"__ignoreMap":575},[2393,2417,2434,2452,2470,2488,2506,2524,2542],{"type":42,"tag":2394,"props":2395,"children":2398},"span",{"class":2396,"line":2397},"line",1,[2399,2405,2411],{"type":42,"tag":2394,"props":2400,"children":2402},{"style":2401},"--shiki-light:#E2931D;--shiki-default:#FFCB6B;--shiki-dark:#FFCB6B",[2403],{"type":48,"value":2404},"pnpm",{"type":42,"tag":2394,"props":2406,"children":2408},{"style":2407},"--shiki-light:#91B859;--shiki-default:#C3E88D;--shiki-dark:#C3E88D",[2409],{"type":48,"value":2410}," dev",{"type":42,"tag":2394,"props":2412,"children":2414},{"style":2413},"--shiki-light:#90A4AE;--shiki-light-font-style:italic;--shiki-default:#546E7A;--shiki-default-font-style:italic;--shiki-dark:#676E95;--shiki-dark-font-style:italic",[2415],{"type":48,"value":2416},"          # local dev server\n",{"type":42,"tag":2394,"props":2418,"children":2419},{"class":2396,"line":29},[2420,2424,2429],{"type":42,"tag":2394,"props":2421,"children":2422},{"style":2401},[2423],{"type":48,"value":2404},{"type":42,"tag":2394,"props":2425,"children":2426},{"style":2407},[2427],{"type":48,"value":2428}," build",{"type":42,"tag":2394,"props":2430,"children":2431},{"style":2413},[2432],{"type":48,"value":2433},"        # production build\n",{"type":42,"tag":2394,"props":2435,"children":2437},{"class":2396,"line":2436},3,[2438,2442,2447],{"type":42,"tag":2394,"props":2439,"children":2440},{"style":2401},[2441],{"type":48,"value":2404},{"type":42,"tag":2394,"props":2443,"children":2444},{"style":2407},[2445],{"type":48,"value":2446}," lint",{"type":42,"tag":2394,"props":2448,"children":2449},{"style":2413},[2450],{"type":48,"value":2451},"         # next lint\n",{"type":42,"tag":2394,"props":2453,"children":2455},{"class":2396,"line":2454},4,[2456,2460,2465],{"type":42,"tag":2394,"props":2457,"children":2458},{"style":2401},[2459],{"type":48,"value":2404},{"type":42,"tag":2394,"props":2461,"children":2462},{"style":2407},[2463],{"type":48,"value":2464}," test",{"type":42,"tag":2394,"props":2466,"children":2467},{"style":2413},[2468],{"type":48,"value":2469},"         # vitest run (needs .env.local with real API keys + DATABASE_URL)\n",{"type":42,"tag":2394,"props":2471,"children":2473},{"class":2396,"line":2472},5,[2474,2478,2483],{"type":42,"tag":2394,"props":2475,"children":2476},{"style":2401},[2477],{"type":48,"value":2404},{"type":42,"tag":2394,"props":2479,"children":2480},{"style":2407},[2481],{"type":48,"value":2482}," test:watch",{"type":42,"tag":2394,"props":2484,"children":2485},{"style":2413},[2486],{"type":48,"value":2487},"   # vitest watch mode\n",{"type":42,"tag":2394,"props":2489,"children":2491},{"class":2396,"line":2490},6,[2492,2496,2501],{"type":42,"tag":2394,"props":2493,"children":2494},{"style":2401},[2495],{"type":48,"value":2404},{"type":42,"tag":2394,"props":2497,"children":2498},{"style":2407},[2499],{"type":48,"value":2500}," db:push",{"type":42,"tag":2394,"props":2502,"children":2503},{"style":2413},[2504],{"type":48,"value":2505},"      # push schema changes to Neon\n",{"type":42,"tag":2394,"props":2507,"children":2509},{"class":2396,"line":2508},7,[2510,2514,2519],{"type":42,"tag":2394,"props":2511,"children":2512},{"style":2401},[2513],{"type":48,"value":2404},{"type":42,"tag":2394,"props":2515,"children":2516},{"style":2407},[2517],{"type":48,"value":2518}," db:generate",{"type":42,"tag":2394,"props":2520,"children":2521},{"style":2413},[2522],{"type":48,"value":2523},"  # generate migration SQL\n",{"type":42,"tag":2394,"props":2525,"children":2527},{"class":2396,"line":2526},8,[2528,2532,2537],{"type":42,"tag":2394,"props":2529,"children":2530},{"style":2401},[2531],{"type":48,"value":2404},{"type":42,"tag":2394,"props":2533,"children":2534},{"style":2407},[2535],{"type":48,"value":2536}," db:studio",{"type":42,"tag":2394,"props":2538,"children":2539},{"style":2413},[2540],{"type":48,"value":2541},"    # Drizzle Studio GUI\n",{"type":42,"tag":2394,"props":2543,"children":2545},{"class":2396,"line":2544},9,[2546,2550,2555],{"type":42,"tag":2394,"props":2547,"children":2548},{"style":2401},[2549],{"type":48,"value":2404},{"type":42,"tag":2394,"props":2551,"children":2552},{"style":2407},[2553],{"type":48,"value":2554}," workflows",{"type":42,"tag":2394,"props":2556,"children":2557},{"style":2413},[2558],{"type":48,"value":2559},"    # inspect workflow runs (npx workflow inspect runs --web)\n",{"type":42,"tag":65,"props":2561,"children":2563},{"id":2562},"environment-variables",[2564],{"type":48,"value":2565},"Environment Variables",{"type":42,"tag":146,"props":2567,"children":2568},{},[2569,2585],{"type":42,"tag":150,"props":2570,"children":2571},{},[2572],{"type":42,"tag":154,"props":2573,"children":2574},{},[2575,2580],{"type":42,"tag":158,"props":2576,"children":2577},{},[2578],{"type":48,"value":2579},"Var",{"type":42,"tag":158,"props":2581,"children":2582},{},[2583],{"type":48,"value":2584},"Purpose",{"type":42,"tag":169,"props":2586,"children":2587},{},[2588,2604,2620,2636,2652,2668,2690],{"type":42,"tag":154,"props":2589,"children":2590},{},[2591,2599],{"type":42,"tag":176,"props":2592,"children":2593},{},[2594],{"type":42,"tag":180,"props":2595,"children":2597},{"className":2596},[],[2598],{"type":48,"value":1035},{"type":42,"tag":176,"props":2600,"children":2601},{},[2602],{"type":48,"value":2603},"Vercel AI Gateway — required for all AI calls",{"type":42,"tag":154,"props":2605,"children":2606},{},[2607,2615],{"type":42,"tag":176,"props":2608,"children":2609},{},[2610],{"type":42,"tag":180,"props":2611,"children":2613},{"className":2612},[],[2614],{"type":48,"value":403},{"type":42,"tag":176,"props":2616,"children":2617},{},[2618],{"type":48,"value":2619},"Exa search API — required for research",{"type":42,"tag":154,"props":2621,"children":2622},{},[2623,2631],{"type":42,"tag":176,"props":2624,"children":2625},{},[2626],{"type":42,"tag":180,"props":2627,"children":2629},{"className":2628},[],[2630],{"type":48,"value":1454},{"type":42,"tag":176,"props":2632,"children":2633},{},[2634],{"type":48,"value":2635},"Neon Postgres connection string",{"type":42,"tag":154,"props":2637,"children":2638},{},[2639,2647],{"type":42,"tag":176,"props":2640,"children":2641},{},[2642],{"type":42,"tag":180,"props":2643,"children":2645},{"className":2644},[],[2646],{"type":48,"value":1194},{"type":42,"tag":176,"props":2648,"children":2649},{},[2650],{"type":48,"value":2651},"Outreach OAuth (optional — only if using Outreach)",{"type":42,"tag":154,"props":2653,"children":2654},{},[2655,2663],{"type":42,"tag":176,"props":2656,"children":2657},{},[2658],{"type":42,"tag":180,"props":2659,"children":2661},{"className":2660},[],[2662],{"type":48,"value":1201},{"type":42,"tag":176,"props":2664,"children":2665},{},[2666],{"type":48,"value":2667},"Outreach OAuth (optional)",{"type":42,"tag":154,"props":2669,"children":2670},{},[2671,2679],{"type":42,"tag":176,"props":2672,"children":2673},{},[2674],{"type":42,"tag":180,"props":2675,"children":2677},{"className":2676},[],[2678],{"type":48,"value":1208},{"type":42,"tag":176,"props":2680,"children":2681},{},[2682,2684],{"type":48,"value":2683},"Must be ",{"type":42,"tag":180,"props":2685,"children":2687},{"className":2686},[],[2688],{"type":48,"value":2689},"https:\u002F\u002F\u003Cdomain>\u002Fapi\u002Foauth\u002Foutreach\u002Fcallback",{"type":42,"tag":154,"props":2691,"children":2692},{},[2693,2702],{"type":42,"tag":176,"props":2694,"children":2695},{},[2696],{"type":42,"tag":180,"props":2697,"children":2699},{"className":2698},[],[2700],{"type":48,"value":2701},"NEXT_PUBLIC_SITE_URL",{"type":42,"tag":176,"props":2703,"children":2704},{},[2705],{"type":48,"value":2706},"Used for the API curl example on campaign detail page",{"type":42,"tag":65,"props":2708,"children":2710},{"id":2709},"answering-questions",[2711],{"type":48,"value":2712},"Answering Questions",{"type":42,"tag":51,"props":2714,"children":2715},{},[2716],{"type":48,"value":2717},"When a user asks \"how do I…\" about this repo:",{"type":42,"tag":91,"props":2719,"children":2720},{},[2721,2726,2731,2736],{"type":42,"tag":95,"props":2722,"children":2723},{},[2724],{"type":48,"value":2725},"Identify which integration\u002Flayer they're asking about (research, AI, CRM, DB, UI, workflow).",{"type":42,"tag":95,"props":2727,"children":2728},{},[2729],{"type":48,"value":2730},"Reference the exact files above so they can navigate directly.",{"type":42,"tag":95,"props":2732,"children":2733},{},[2734],{"type":48,"value":2735},"If they want to swap an integration, point them to the \"How To Extend\" section that matches and name the specific files to change.",{"type":42,"tag":95,"props":2737,"children":2738},{},[2739],{"type":48,"value":2740},"If unsure about current behavior, read the specific file before answering — this guide is a map, not a substitute for the code.",{"type":42,"tag":2742,"props":2743,"children":2744},"style",{},[2745],{"type":48,"value":2746},"html .light .shiki span {color: var(--shiki-light);background: var(--shiki-light-bg);font-style: var(--shiki-light-font-style);font-weight: var(--shiki-light-font-weight);text-decoration: var(--shiki-light-text-decoration);}html.light .shiki span {color: var(--shiki-light);background: var(--shiki-light-bg);font-style: var(--shiki-light-font-style);font-weight: var(--shiki-light-font-weight);text-decoration: var(--shiki-light-text-decoration);}html .default .shiki span {color: var(--shiki-default);background: var(--shiki-default-bg);font-style: var(--shiki-default-font-style);font-weight: var(--shiki-default-font-weight);text-decoration: var(--shiki-default-text-decoration);}html .shiki span {color: var(--shiki-default);background: var(--shiki-default-bg);font-style: var(--shiki-default-font-style);font-weight: var(--shiki-default-font-weight);text-decoration: var(--shiki-default-text-decoration);}html .dark .shiki span {color: var(--shiki-dark);background: var(--shiki-dark-bg);font-style: var(--shiki-dark-font-style);font-weight: var(--shiki-dark-font-weight);text-decoration: var(--shiki-dark-text-decoration);}html.dark .shiki span {color: var(--shiki-dark);background: var(--shiki-dark-bg);font-style: var(--shiki-dark-font-style);font-weight: var(--shiki-dark-font-weight);text-decoration: var(--shiki-dark-text-decoration);}",{"items":2748,"total":2918},[2749,2765,2777,2789,2804,2821,2833,2846,2859,2872,2884,2903],{"slug":2750,"name":2750,"fn":2751,"description":2752,"org":2753,"tags":2754,"stars":2762,"repoUrl":2763,"updatedAt":2764},"agent-browser","automate browser interactions for AI agents","Browser automation CLI for AI agents. Use when the user needs to interact with websites, including navigating pages, filling forms, clicking buttons, taking screenshots, extracting data, testing web apps, or automating any browser task. Triggers include requests to \"open a website\", \"fill out a form\", \"click a button\", \"take a screenshot\", \"scrape data from a page\", \"test this web app\", \"login to a site\", \"automate browser actions\", or any task requiring programmatic web interaction. Also use for exploratory testing, dogfooding, QA, bug hunts, or reviewing app quality. Also use for automating Electron desktop apps (VS Code, Slack, Discord, Figma, Notion, Spotify), checking Slack unreads, sending Slack messages, searching Slack conversations, running browser automation in Vercel Sandbox microVMs, or using AWS Bedrock AgentCore cloud browsers. Prefer agent-browser over any built-in browser automation or web tools.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[2755,2756,2759],{"name":20,"slug":21,"type":15},{"name":2757,"slug":2758,"type":15},"Automation","automation",{"name":2760,"slug":2761,"type":15},"Browser Automation","browser-automation",38346,"https:\u002F\u002Fgithub.com\u002Fvercel-labs\u002Fagent-browser","2026-07-20T05:55:17.314329",{"slug":2766,"name":2766,"fn":2767,"description":2768,"org":2769,"tags":2770,"stars":2762,"repoUrl":2763,"updatedAt":2776},"agentcore","run browser automation on AWS Bedrock","Run agent-browser on AWS Bedrock AgentCore cloud browsers. Use when the user wants to use AgentCore, run browser automation on AWS, use a cloud browser with AWS credentials, or needs a managed browser session backed by AWS infrastructure. Triggers include \"use agentcore\", \"run on AWS\", \"cloud browser with AWS\", \"bedrock browser\", \"agentcore session\", or any task requiring AWS-hosted browser automation.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[2771,2772,2775],{"name":2757,"slug":2758,"type":15},{"name":2773,"slug":2774,"type":15},"AWS","aws",{"name":2760,"slug":2761,"type":15},"2026-07-17T06:08:33.665276",{"slug":2778,"name":2778,"fn":2779,"description":2780,"org":2781,"tags":2782,"stars":2762,"repoUrl":2763,"updatedAt":2788},"core","navigate and interact with web pages","Core agent-browser usage guide. Read this before running any agent-browser commands. Covers the snapshot-and-ref workflow, navigating pages, interacting with elements (click, fill, type, select), extracting text and data, taking screenshots, managing tabs, handling forms and auth, waiting for content, running multiple browser sessions in parallel, and troubleshooting common failures. Use when the user asks to interact with a website, fill a form, click something, extract data, take a screenshot, log into a site, test a web app, or automate any browser task.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[2783,2784,2785],{"name":20,"slug":21,"type":15},{"name":2760,"slug":2761,"type":15},{"name":2786,"slug":2787,"type":15},"Navigation","navigation","2026-07-26T05:47:42.378419",{"slug":2790,"name":2790,"fn":2791,"description":2792,"org":2793,"tags":2794,"stars":2762,"repoUrl":2763,"updatedAt":2803},"derive-client","reverse engineer internal APIs from browser traffic","Reverse-engineer a website's internal API by recording browser traffic into a HAR file, then generate a standalone client or CLI that calls the endpoints directly, with no browser needed after the first recording. Use when asked to \"derive a client\", \"build a CLI for \u003Csite>\", \"reverse engineer this site's API\", \"record network requests\", \"turn this site into an API\", or when the same site will be automated repeatedly and direct HTTP calls would beat driving the browser every time.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[2795,2798,2799,2800],{"name":2796,"slug":2797,"type":15},"API Development","api-development",{"name":2757,"slug":2758,"type":15},{"name":2760,"slug":2761,"type":15},{"name":2801,"slug":2802,"type":15},"Web Scraping","web-scraping","2026-07-20T06:24:11.928835",{"slug":2805,"name":2805,"fn":2806,"description":2807,"org":2808,"tags":2809,"stars":2762,"repoUrl":2763,"updatedAt":2820},"dogfood","perform exploratory testing on web applications","Systematically explore and test a web application to find bugs, UX issues, and other problems. Use when asked to \"dogfood\", \"QA\", \"exploratory test\", \"find issues\", \"bug hunt\", \"test this app\u002Fsite\u002Fplatform\", or review the quality of a web application. Produces a structured report with full reproduction evidence -- step-by-step screenshots, repro videos, and detailed repro steps for every issue -- so findings can be handed directly to the responsible teams.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[2810,2811,2814,2817],{"name":2760,"slug":2761,"type":15},{"name":2812,"slug":2813,"type":15},"Debugging","debugging",{"name":2815,"slug":2816,"type":15},"QA","qa",{"name":2818,"slug":2819,"type":15},"Testing","testing","2026-07-17T06:07:41.421482",{"slug":2822,"name":2822,"fn":2823,"description":2824,"org":2825,"tags":2826,"stars":2762,"repoUrl":2763,"updatedAt":2832},"electron","automate Electron desktop applications","Automate Electron desktop apps (VS Code, Slack, Discord, Figma, Notion, Spotify, etc.) using agent-browser via Chrome DevTools Protocol. Use when the user needs to interact with an Electron app, automate a desktop app, connect to a running app, control a native app, or test an Electron application. Triggers include \"automate Slack app\", \"control VS Code\", \"interact with Discord app\", \"test this Electron app\", \"connect to desktop app\", or any task requiring automation of a native Electron application.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[2827,2828,2829],{"name":20,"slug":21,"type":15},{"name":2760,"slug":2761,"type":15},{"name":2830,"slug":2831,"type":15},"Desktop","desktop","2026-07-17T06:08:28.007783",{"slug":2834,"name":2834,"fn":2835,"description":2836,"org":2837,"tags":2838,"stars":2762,"repoUrl":2763,"updatedAt":2845},"slack","interact with Slack workspaces","Interact with Slack workspaces using browser automation. Use when the user needs to check unread channels, navigate Slack, send messages, extract data, find information, search conversations, or automate any Slack task. Triggers include \"check my Slack\", \"what channels have unreads\", \"send a message to\", \"search Slack for\", \"extract from Slack\", \"find who said\", or any task requiring programmatic Slack interaction.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[2839,2840,2843],{"name":2760,"slug":2761,"type":15},{"name":2841,"slug":2842,"type":15},"Messaging","messaging",{"name":2844,"slug":2834,"type":15},"Slack","2026-07-17T06:08:27.679015",{"slug":2847,"name":2847,"fn":2848,"description":2849,"org":2850,"tags":2851,"stars":2762,"repoUrl":2763,"updatedAt":2858},"vercel-sandbox","run browser automation in Vercel Sandbox","Run agent-browser + Chrome inside Vercel Sandbox microVMs for browser automation from any Vercel-deployed app. Use when the user needs browser automation in a Vercel app (Next.js, SvelteKit, Nuxt, Remix, Astro, etc.), wants to run headless Chrome without binary size limits, needs persistent browser sessions across commands, or wants ephemeral isolated browser environments. Triggers include \"Vercel Sandbox browser\", \"microVM Chrome\", \"agent-browser in sandbox\", \"browser automation on Vercel\", or any task requiring Chrome in a Vercel Sandbox.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[2852,2853,2854,2855],{"name":2757,"slug":2758,"type":15},{"name":2760,"slug":2761,"type":15},{"name":2818,"slug":2819,"type":15},{"name":2856,"slug":2857,"type":15},"Vercel","vercel","2026-07-17T06:08:28.349899",{"slug":2860,"name":2860,"fn":2861,"description":2862,"org":2863,"tags":2864,"stars":2869,"repoUrl":2870,"updatedAt":2871},"deploy-to-vercel","deploy applications to Vercel","Deploy applications and websites to Vercel. Use when the user requests deployment actions like \"deploy my app\", \"deploy and give me the link\", \"push this live\", or \"create a preview deployment\".",{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[2865,2868],{"name":2866,"slug":2867,"type":15},"Deployment","deployment",{"name":2856,"slug":2857,"type":15},28993,"https:\u002F\u002Fgithub.com\u002Fvercel-labs\u002Fagent-skills","2026-07-17T06:08:41.18374",{"slug":2873,"name":2873,"fn":2874,"description":2875,"org":2876,"tags":2877,"stars":2869,"repoUrl":2870,"updatedAt":2883},"vercel-cli-with-tokens","manage Vercel projects via CLI","Deploy and manage projects on Vercel using token-based authentication. Use when working with Vercel CLI using access tokens rather than interactive login — e.g. \"deploy to vercel\", \"set up vercel\", \"add environment variables to vercel\".",{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[2878,2881,2882],{"name":2879,"slug":2880,"type":15},"CLI","cli",{"name":2866,"slug":2867,"type":15},{"name":2856,"slug":2857,"type":15},"2026-07-17T06:08:41.84179",{"slug":2885,"name":2885,"fn":2886,"description":2887,"org":2888,"tags":2889,"stars":2869,"repoUrl":2870,"updatedAt":2902},"vercel-composition-patterns","implement scalable React composition patterns","React composition patterns that scale. Use when refactoring components with boolean prop proliferation, building flexible component libraries, or designing reusable APIs. Triggers on tasks involving compound components, render props, context providers, or component architecture. Includes React 19 API changes.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[2890,2893,2896,2899],{"name":2891,"slug":2892,"type":15},"Best Practices","best-practices",{"name":2894,"slug":2895,"type":15},"Frontend","frontend",{"name":2897,"slug":2898,"type":15},"React","react",{"name":2900,"slug":2901,"type":15},"UI Components","ui-components","2026-07-17T06:05:40.576913",{"slug":2904,"name":2904,"fn":2905,"description":2906,"org":2907,"tags":2908,"stars":2869,"repoUrl":2870,"updatedAt":2917},"vercel-optimize","optimize Vercel project performance and costs","Use for Vercel cost and performance optimization on deployed projects, especially Next.js, SvelteKit, Nuxt, and limited Astro apps. Collect Vercel metrics, usage, project config, and code scan results first; investigate only metric-backed candidates; produce ranked recommendations grounded in verified files and version-aware Vercel\u002Fframework docs. Trigger for Vercel bill reduction, slow or expensive routes, caching opportunities, Function Invocations, Build Minutes, Fast Data Transfer, Core Web Vitals, Bot Management, Fluid compute, or cost breakdown requests.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[2909,2912,2913,2916],{"name":2910,"slug":2911,"type":15},"Cost Optimization","cost-optimization",{"name":2866,"slug":2867,"type":15},{"name":2914,"slug":2915,"type":15},"Performance","performance",{"name":2856,"slug":2857,"type":15},"2026-07-17T06:04:08.327515",100,{"items":2920,"total":2397},[2921],{"slug":4,"name":4,"fn":5,"description":6,"org":2922,"tags":2923,"stars":25,"repoUrl":26,"updatedAt":27},{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[2924,2925,2926,2927],{"name":20,"slug":21,"type":15},{"name":17,"slug":18,"type":15},{"name":23,"slug":24,"type":15},{"name":13,"slug":14,"type":15}]