> For the complete documentation index, see [llms.txt](https://docs.finchagentic.com/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.finchagentic.com/readme.md).

# Finch Agentic

**Finch** is the runtime layer for Agentic AI - persistent memory, vault, agents, workflows, and execution tools that survive every session.

This documentation covers:

1. **Finch MCP** (`@finchagentic/mcp`) — install once into Claude Code, Cursor, Hermes, Windsurf, Codex, and any MCP client
2. **Finch App** — Terminal, Trading, Wallet, API Market, Skills in the browser

> npm package: `@finchagentic/mcp` · product brand: **Finch**.

***

## Start here

| Path                                   | What                                          |
| -------------------------------------- | --------------------------------------------- |
| [Getting started](/getting-started.md) | 60-second path from zero to first memory save |
| [Install](/install.md)                 | Every MCP client + pin rules                  |
| [Local mode](/local-mode.md)           | Vault + memory on your machine, no account    |
| [Configuration](/configuration.md)     | Env vars, tool palettes, providers            |
| [CLI](/cli.md)                         | `finch` / doctor / setup / vault              |

## Product pillars

| Pillar        | Docs                                                                     |
| ------------- | ------------------------------------------------------------------------ |
| **Memory**    | [memory.md](/memory.md) — semantic recall that auto-loads                |
| **Vault**     | [vault.md](/vault.md) — versioned notes, graph, credentials              |
| **Agents**    | [agents.md](/agents.md) — spawn, schedule, audit ledger                  |
| **Workflows** | [workflows.md](/workflows.md) — automations, monitors, packets, research |

## Tools reference

| Doc                                            | Scope                              |
| ---------------------------------------------- | ---------------------------------- |
| [Tools overview](/tools/overview.md)           | 121 tools · categories · palettes  |
| [Memory & vault tools](/tools/memory-vault.md) | `memory_*` · `vault_*` · chronicle |
| [Agent tools](/tools/agents.md)                | lifecycle · identity · ledger      |
| [Base DeFi](/tools/base-defi.md)               | `base_mcp_*` swaps, send, lend     |
| [Robinhood Chain](/tools/robinhood-chain.md)   | `rh_mcp_*` stocks + crypto         |
| [Research](/tools/research.md)                 | web · deep\_research · compare     |
| [Market & scanner](/tools/market.md)           | boards · scores · playbooks        |
| [GitHub & OS](/tools/github-os.md)             | github\_\* · diagnostics · shell   |

## Finch App

| Doc                                        | Scope                                            |
| ------------------------------------------ | ------------------------------------------------ |
| [App overview](/app/overview.md)           | Surfaces, auth, stack                            |
| [Terminal](/app/terminal.md)               | Chat tool-calling (analytics, swap, automations) |
| [Trading & wallet](/app/trading-wallet.md) | Base + RH Chain execution wallet                 |
| [API Market](/app/api-market.md)           | Buy / Sell inference · OpenAI-compatible         |
| [Skills](/app/skills.md)                   | Skill marketplace                                |

## Safety & ops

| Doc                                    | Scope                                     |
| -------------------------------------- | ----------------------------------------- |
| [Security](/security.md)               | 8 mandatory boundaries                    |
| [Troubleshooting](/troubleshooting.md) | doctor, common failures                   |
| [Clients](/clients.md)                 | Claude Code, Cursor, Hermes, VS Code, Zed |

***

## Quick facts

| Fact                          | Value                                             |
| ----------------------------- | ------------------------------------------------- |
| MCP package                   | `@finchagentic/mcp@4.0.0` (pin — never `@latest`) |
| Registered tools              | **121**                                           |
| Default tool palette          | `core` (set `FINCH_TOOLS=all` for full set)       |
| Primary chains                | Base `8453` · Robinhood Chain `4663`              |
| Settlement asset (API Market) | USDG on Robinhood Chain                           |
| npm                           | <https://www.npmjs.com/package/@finchagentic/mcp> |

***

## What Finch is *not*

* Not a generic “121 tools dump” pitch without persistence — the product is **memory + agents that stick**
* Not Robinhood brokerage / `agent.robinhood.com` — RH tools are **Robinhood Chain** (on-chain tokenized stocks + crypto)
* Not unauthenticated mainnet spend — swaps/sends need estimate → preview → confirm


---

# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.

## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation by asking a question.

Perform an HTTP GET request on the following URL with the `ask` and `goal` query parameters:

```
GET https://docs.finchagentic.com/readme.md?ask=<question>&goal=<user_goal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is what the user is ultimately trying to achieve, the reason they need the answer. Sharing it helps GitBook give you a better, more relevant answer. A goal is most helpful when it describes the outcome the user wants rather than restating the question. For example, with `ask=how do I create an API token`, a goal like `automate deployments from our CI pipeline` lets GitBook tailor the answer to that use case.

The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.