# Alfera

## Docs

- [Alfera Documentation](https://docs.alfera.ai/index.md): Alfera is the AI employee that lives in Slack. Describe a task and it runs end to end across your connected tools.
- [Get Started with Alfera in Minutes](https://docs.alfera.ai/quickstart.md): Connect your first tool, assign your first task, and see Alfera execute it end to end in plain English. No code or setup wizards required.
- [How Alfera Works: From Task to Finished Output](https://docs.alfera.ai/how-it-works.md): Learn how Alfera reads your request, plans the steps, executes across your connected tools, and hands back a finished result with a full audit trail.
- [Customer Support AI Employee: Triage and Reply at Scale](https://docs.alfera.ai/use-cases/customer-support.md): Use Alfera to triage your support inbox, draft and send routine replies, confirm appointments, and escalate only the issues that need a human.
- [Sales Follow-Up AI Employee: Chase Every Lead Automatically](https://docs.alfera.ai/use-cases/sales-follow-up.md): Use Alfera to follow up on stale quotes, update CRM records, enrich contacts, draft proposals, and book meetings without manual effort.
- [Operations AI Employee: Run Recurring Work Without You](https://docs.alfera.ai/use-cases/operations-admin.md): Use Alfera to keep spreadsheets clean, prepare weekly owner reports, chase missing inputs, and sync data across your tools on a schedule.
- [Bookkeeping AI Employee: Chase Invoices and Close the Books](https://docs.alfera.ai/use-cases/bookkeeping.md): Use Alfera to detect overdue invoices, send on-brand payment reminders, match payments, flag mismatches, and organize documents for month-end.
- [Scheduling AI Employee: Fewer No-Shows, Zero Phone Tag](https://docs.alfera.ai/use-cases/scheduling.md): Use Alfera to send appointment confirmations and reminders, reschedule cleanly, and keep team calendars balanced and up to date.
- [Ecommerce Support AI Employee: Resolve Orders at Scale](https://docs.alfera.ai/use-cases/ecommerce-support.md): Use Alfera to answer order status questions, flag and prep refund requests, escalate edge cases, and summarize support themes for your store.
- [Alfera Integrations: Connect 3,000+ Tools in One Click](https://docs.alfera.ai/integrations/overview.md): Alfera plugs into every tool your business already uses: email, CRM, payments, spreadsheets, scheduling, your store, and thousands more via secure OAuth.
- [How to Connect a Tool to Your Alfera Workspace](https://docs.alfera.ai/integrations/connect-a-tool.md): Add a new integration to Alfera in a few clicks using secure OAuth. Connections are scoped, encrypted, and revocable from your workspace settings.
- [How to Add an MCP Server to Your Alfera Workspace](https://docs.alfera.ai/integrations/add-an-mcp-server.md): Connect any MCP server to Alfera by pasting its address. Its tools become available to your workers, with the same approvals, encryption, and per-worker controls as every other integration.
- [Approval Gates: Stay in Control of Every Sensitive Action](https://docs.alfera.ai/control/approvals.md): Alfera pauses before sending to a customer or changing important data, and waits for your one-tap approval. Set what runs automatically and what needs review.
- [Audit Trail: A Plain-English Record of Every Run](https://docs.alfera.ai/control/audit-trail.md): Every Alfera run produces a readable log of what it read, changed, sent, and skipped so you can review and reverse any action at any time.
- [Security at Alfera: How Your Data Is Protected](https://docs.alfera.ai/control/security.md): Alfera encrypts all credentials with AES-256-GCM, never shares or sells your data, never trains models on it, and gives you full control over every connection.
- [Alfera plans: Free, Pro, and Enterprise](https://docs.alfera.ai/billing/plans.md): Start free with no credit card. Upgrade to Pro at $50 per month for recurring work. Enterprise adds custom data regions, SSO, and dedicated support.
- [Credits: how Alfera charges for work](https://docs.alfera.ai/billing/credits.md): Alfera runs on credits. One dollar is 100 credits. You spend credits only when a task runs — never for seats, integrations, or idle time.
- [Alfera Brain API: Workspace Memory and AI Answers](https://docs.alfera.ai/api-reference/overview.md): The Brain API gives your product access to your Alfera workspace memory over plain HTTP. Search facts, build context packets, teach new facts, or get AI answers.
- [Brain API Authentication: API Keys, Scopes, and Audiences](https://docs.alfera.ai/api-reference/authentication.md): Create an API key in Settings, choose its scopes (memory.read, memory.write, answers.write) and audience (internal or external), then pass it as a Bearer token.
- [POST /v1/memory/search: Search Workspace Memory](https://docs.alfera.ai/api-reference/memory-search.md): Search the workspace brain and return ranked matching facts. Scope: memory.read. Pass end_user.id to include facts about a specific person.
- [POST /v1/memory/context: Build a Memory Context Packet](https://docs.alfera.ai/api-reference/memory-context.md): Returns the rendered memory block the worker gets each turn, inside a token budget. Scope: memory.read. Ready to paste into your own model prompt.
- [POST /v1/memory/facts: Teach the Brain New Facts](https://docs.alfera.ai/api-reference/memory-facts.md): Submit new facts to the workspace brain via the same capture pipeline the worker uses. Scope: memory.write. Replaying the same call is safe; duplicates are skipped.
- [POST /v1/answers: Ask the Worker and Stream or Poll a Reply](https://docs.alfera.ai/api-reference/answers.md): Run the workspace worker on a question and stream the reply as server-sent events, or queue it and poll by answer_id. Scope: answers.write.
- [Brain MCP Server: workspace memory as tools](https://docs.alfera.ai/api-reference/mcp/overview.md): Mount your Alfera workspace brain as tools in Claude Code, Cursor, or your own agent via the MCP server at https://api.alfera.ai/v1/mcp.
- [Brain MCP tools: reference for all available tools](https://docs.alfera.ai/api-reference/mcp/tools.md): Full reference for the six Brain MCP tools: whoami, search_memory, get_memory_context, teach_memory, ask_worker, and wait_for_answer.