# Concierca AI > Concierca AI documentation — user guide and MCP API reference. Configure AI agents on your company's data: missions, knowledge, memory, skills and integrations. ## Docs - [Your First 10 Minutes](https://agents.concierca.ch/guide/first-run.md): The welcome wizard: company profile, departments, your AI provider key, and your first agent — from empty workspace to first agent run. - [Introduction](https://agents.concierca.ch/index.md): Welcome to your project - [Overview & Navigation](https://agents.concierca.ch/guide/introduction.md): What Concierca AI is, the app shell, sidebar, search, departments, roles, and credits. - [Quickstart](https://agents.concierca.ch/quickstart.md): Begin with a guide on the fastest path to a successful outcome - [Chat](https://agents.concierca.ch/guide/chat.md): Talking to your company assistant: conversations, context, slash commands, and the Data Assistant. - [Inbox](https://agents.concierca.ch/guide/inbox.md): Your company inbox: messages, questions, and escalations from agents — and how to answer them. - [Company Hub](https://agents.concierca.ch/guide/company-hub.md): Strategy, SOP, and Decisions views, and the admin-only Brain page. - [Agents](https://agents.concierca.ch/guide/agents.md): Creating and configuring agents: briefing, skills, tools, data access, triggers, and the hierarchy. - [Hierarchy](https://agents.concierca.ch/guide/agent-hierarchy.md): The agent org chart and the delegation protocol: reports-to chains, handoffs, and delegated runs. - [Runs](https://agents.concierca.ch/guide/agent-runs.md): The live table of every agent execution, plus per-run detail. - [Agent Decision Log](https://agents.concierca.ch/guide/agent-decision-log.md): The audit trail of agent decisions, tool calls, escalations, and self-heal events. - [Agent Inbox](https://agents.concierca.ch/guide/agent-inbox.md): The agent-to-agent message log: handoffs, delegated tasks, and reports up the chain. - [Analytics](https://agents.concierca.ch/guide/agent-analytics.md): A dashboard of agent cost, performance, and reliability. - [Cost](https://agents.concierca.ch/guide/agent-cost.md): The Agent Cost Dashboard: budget, spend by agent, and the most expensive runs. - [Missions](https://agents.concierca.ch/guide/missions.md): Multi-step agent workflows: blueprints, runs, artifacts, approvals, and budgets. - [Knowledge](https://agents.concierca.ch/guide/knowledge.md): The company knowledge base: entries, categories, importance, and visibility. - [Memories](https://agents.concierca.ch/guide/memories.md): What agents remember: saving, editing, and scoping memories. - [Skills & Skills Review](https://agents.concierca.ch/guide/skills-sops.md): Reusable how-to instructions for agents, and the admin Skills Review queue. - [Templates & Marketplace](https://agents.concierca.ch/guide/templates.md): Knowledge templates and the agent template marketplace (admin). - [How Your Workspace Learns](https://agents.concierca.ch/guide/learning.md): The nightly curator: how repeated successes and failures become proposed skills and knowledge — always with your approval. - [Your Data](https://agents.concierca.ch/guide/data.md): A plain-language map of every data domain in your workspace — what each one holds and where to find its details. - [Data Views](https://agents.concierca.ch/guide/data-views.md): Table views over your business data: filtering, columns, editing, and the visibility picker. - [Purchasing & Procurement](https://agents.concierca.ch/guide/procurement.md): How the procurement system turns sales history into one clear buying decision per SKU: the Procurement Queue, Purchase Orders, Suppliers, and the supporting analytics views. - [Automate with AI (MCP)](https://agents.concierca.ch/guide/automate-with-ai.md): Your entire workspace is operable by AI: connect Claude, ChatGPT, or any MCP client and give it 136 governed tools across agents, missions, data, knowledge, and market intelligence. - [Connect via MCP](https://agents.concierca.ch/guide/mcp-connect.md): The human setup guide: what MCP is, your workspace endpoint, connecting Claude and ChatGPT, getting a key, and your first calls. - [Every MCP Tool](https://agents.concierca.ch/guide/mcp-tools.md): The full catalog of all 136 tools your workspace exposes over MCP, grouped by resource with a one-line purpose each. - [CLI](https://agents.concierca.ch/guide/mcp-cli.md): The concierca terminal client: install-free invocation, MCP key setup, environment variables, and the full command set. - [Connections](https://agents.concierca.ch/guide/connections.md): Connecting external tools: OAuth integrations, per-tool controls, and MCP access. - [Approvals](https://agents.concierca.ch/guide/approvals.md): Reviewing and resolving actions that agents need your sign-off for. - [Settings](https://agents.concierca.ch/guide/settings.md): Personal and company settings: account, preferences, users, departments, branding, billing, and monitoring. - [FAQ & Troubleshooting](https://agents.concierca.ch/guide/faq.md): Frequently asked questions, fixes for typical problems, and how to get help. - [Operating Contract](https://agents.concierca.ch/api/ai/operating-contract.md): The session-start contract every MCP client should follow: ground, discover, then act. - [What are you trying to do?](https://agents.concierca.ch/api/ai/routing.md): Use-case-first routing: map an intent to the exact tool sequence. - [Writes, dry-run & readable≠writable](https://agents.concierca.ch/api/ai/writes-and-safety.md): How to write safely: dry-run first, replacement vs additive fields, and the human gate. - [Mission Recipes](https://agents.concierca.ch/api/ai/mission-recipes.md): Operating recipes for missions: explain, start safely, diagnose, control, edit, and compare. - [Vertical Playbook](https://agents.concierca.ch/api/ai/vertical-playbook.md): How an AI uses the market-intelligence layer: start with the legend, route by question, ground every answer. - [Introduction](https://agents.concierca.ch/api/introduction.md): What the MCP server is, endpoints, supported clients, and versioning. - [All Tools](https://agents.concierca.ch/api/tools.md): Every tool the Concierca AI MCP exposes — 136 in total — grouped by resource, with a one-line purpose and a link to full detail. - [Authentication](https://agents.concierca.ch/api/authentication.md): OAuth connect flow, API keys, per-tool scopes, admin vs. member surface, and reconnect behavior. - [Conventions](https://agents.concierca.ch/api/conventions.md): Tool naming, session bootstrap, common parameters, errors, limits, dry-run, and router mode. - [Connect via MCP](https://agents.concierca.ch/api/mcp.md): Connect the Concierca AI MCP server from Claude, ChatGPT, or any MCP client — endpoint, authentication, and client setup. - [Ingest API](https://agents.concierca.ch/api/ingest.md): Push external data into your workspace: API keys, endpoint pattern, idempotent upserts, errors, and Make.com templates. - [CLI](https://agents.concierca.ch/api/cli.md): The concierca terminal client: install-free invocation, MCP key setup, and the v1 command set. - [Decisions](https://agents.concierca.ch/api/data-model/decisions.md): Process decisions and bottleneck interviews — one row per question, from the situation through the human response and the rule it becomes. Ingest-enabled. - [Strategy](https://agents.concierca.ch/api/data-model/strategy.md): Company strategy documents — the knowledge base filtered to category=strategy. - [SOP](https://agents.concierca.ch/api/data-model/sop.md): Standard operating procedures — the knowledge base filtered to category=sop. - [Data](https://agents.concierca.ch/api/data.md): The registry-driven view/table model, data_view_list, data_query, row writes, and view management. - [Row Write Tools](https://agents.concierca.ch/api/write-tools.md): The per-table insert_