Finch

io.github.finchagentic由 GitHub 帳號 finchagentic 發布v4.7.0更新於 Oct 24, 2026

Agentic AI runtime: persistent memory, vault, autonomous agents, deep research, DeFi execution.

已驗證Streamable HTTP可網頁執行金融知識與記憶 AI 與機器學習

服務詳細資訊

識別碼: io.github.finchagentic/mcp
傳輸: Streamable HTTP
執行環境: Web executable
信任: Verified
發布於: Sep 16, 2026

安全提示

MCP 服務會接收對話中的工具呼叫請求，並可能執行外部操作。啟用前請先了解該服務能做什麼，並只授予它所需的權限。

註冊中心已驗證此條目所屬的 GitHub 帳號 finchagentic。這並不表明該帳號就是產品方。

SourceWeft 的檢查只涵蓋條目是否完整、端點是否回應；不審查伺服器的程式碼，也不審查其工具的行為。

在 README 中新增徽章

向讀者展示該 MCP 服務已收錄於 SourceWeft，徽章連結到本頁。

[![Listed on SourceWeft](https://sourceweft.com/badge/mcp/io.github.finchagentic%2Fmcp.svg)](https://sourceweft.com/mcp/io.github.finchagentic%2Fmcp?utm_source=readme&utm_medium=badge)

概覽

AI 產生的概覽

Finch 是代理式 AI 執行環境的 MCP 伺服器，讓助理擁有持久記憶、版本化知識庫、具名代理、工作流程與 DeFi 執行工具。

可以做什麼: 在記憶與知識庫中儲存、全文檢索並版本化知識，並保留歷史紀錄
建立、召回與更新具名代理，並帶入與其目標相關的記憶與知識庫內容
執行自動化、監控、資料包與深度研究，報告會自動儲存並連結過往研究
將編碼與除錯工作階段存成依專案版本化的知識庫條目
執行 Base 與 Robinhood Chain 的 DeFi 操作，並取得行情、網頁與 GitHub 結果
適用情境: 適合希望上下文能跨工作階段保留，或需要代理維持狀態、執行排程研究、進行鏈上操作的場合。適用於需要持久記憶而非一次性工具呼叫的研究、編碼與 DeFi 工作流程。
執行需求: 可作為遠端端點，或以本機 npm 套件（透過 npx 執行，需 Node.js）以 stdio 方式執行；README 建議固定版本。選用帳戶憑證 FINCH_SESSION_TOKEN 或 FINCH_API_KEY；ask_finch 與 deep_research 需一個 LLM 供應商金鑰（BANKR_API_KEY、ANTHROPIC_API_KEY、OPENAI_API_KEY 或 GROK_API_KEY）。選用 FIRECRAWL_API_KEY、GITHUB_TOKEN、ALCHEMY_API_KEY 與 FINCH_WALLET_PASSPHRASE。

安裝前請注意

工具可在鏈上轉移資金（質押、解質押、轉帳、兌換），並會寫入記憶、知識庫與代理；README 指出涉及資金的呼叫需 confirm: true。會建立本機錢包檔案，若未設定 FINCH_WALLET_PASSPHRASE，其金鑰由每次安裝的隨機密鑰與機器資訊衍生。資料可能傳送至 LLM 供應商、Firecrawl、GitHub 與鏈上 RPC。請勿在提示中貼上密鑰，並固定套件版本。

安裝

在 SourceWeft 中

開啟儀表板中的 Finch，將其新增到工作區。
為需要使用其工具的對話啟用該服務。

Web executable，透過 Streamable HTTP。遠端服務在工作區中設定後即可從網頁執行環境執行。

交給代理程式安裝

把這段提示詞貼上到 Claude Code、Codex、Cursor 等代理程式，或 SourceWeft 對話中。代理程式會讀取該伺服器的安裝指南，先向你展示發布者、端點和登入方式，經你同意後再把它新增到你的客戶端。金鑰不會寫進任何命令。

閱讀 https://sourceweft.com/mcp/io.github.finchagentic%2Fmcp/install.md，並按其中說明把這個 MCP 伺服器接入我的客戶端。修改任何設定之前，先告訴我它的發布者、端點和登入方式，等我確認。不要把 API 金鑰寫進命令或我會提交的檔案；問我金鑰應從哪裡取得。

檢視代理程式所依據的安裝指南

其他 MCP 客戶端

把它新增到你客戶端的 mcpServers 設定中。

{
  "mcpServers": {
    "finch": {
      "type": "http",
      "url": "https://mcp.finchagentic.com/mcp"
    }
  }
}

README

Finch

The runtime layer for Agentic AI.

Persistent memory, autonomous agents, and workflows that survive every session.

Docs · App · npm · GitHub · X

Most AI assistants disappear when the conversation ends. Finch gives them lasting state — memory that accumulates, agents that keep running, vaults that version knowledge, and workflows that continue after you close the chat.

Why Finch

	Without Finch	With Finch
Memory	Resets every session	Full-text searchable + versioned vault, decays stale notes
Agents	One-shot tool calls	Named agents with state and audit history
Workflows	Manual chaining	Automations, monitors, packets, deep research
Local	Cloud-only	Vault + memory can run fully on your machine

Install

Always pin the version. Never use @latest.

bash

# One-command installer (detects common MCP clients)npx -y -p @finchagentic/[email protected] finch install

# One-command installer (detects common MCP clients)npx -y -p @finchagentic/[email protected] finch install

Claude Code

Cursor / Windsurf / Claude Desktop

json

{  "mcpServers": {    "finch": {      "command": "npx",      "args": ["-y", "-p", "@finchagentic/[email protected]", "finch-mcp"]    }  }}

{  "mcpServers": {    "finch": {      "command": "npx",      "args": ["-y", "-p", "@finchagentic/[email protected]", "finch-mcp"]    }  }}

VS Code

json

{  "servers": {    "finch": {      "type": "stdio",      "command": "npx",      "args": ["-y", "-p", "@finchagentic/[email protected]", "finch-mcp"]    }  }}

{  "servers": {    "finch": {      "type": "stdio",      "command": "npx",      "args": ["-y", "-p", "@finchagentic/[email protected]", "finch-mcp"]    }  }}

Config file paths

Client	Path
Claude Desktop (Mac)	`~/Library/Application Support/Claude/claude_desktop_config.json`
Claude Desktop (Windows)	`%APPDATA%\Claude\claude_desktop_config.json`
Cursor	`.cursor/mcp.json`
Windsurf	`~/.codeium/windsurf/mcp_config.json`
VS Code	`.vscode/mcp.json`
Zed	`.config/zed/settings.json`

No LLM API key is required to start — 111 of 116 tools are plain reads/writes/on-chain calls that your MCP client's own model already drives; only 5 (ask_finch, deep_research, and scheduled agent learning) do their own multi-step reasoning server-side and need a key (see Configuration). Tools load on first use.

Quick start

bash

finch doctor          # health checkfinch setup           # local vault / memory / providersfinch vault           # inspect local vaultfinch orders          # schedule Robinhood Chain DCA / TP-SL order ticks

finch doctor          # health checkfinch setup           # local vault / memory / providersfinch vault           # inspect local vaultfinch orders          # schedule Robinhood Chain DCA / TP-SL order ticks

Try in your MCP client:

remember: I prefer conservative DeFi strategies, max 5% riskspawn an agent called research-bot to track AI agent news, update it after each sessionsave this thesis to vault

remember: I prefer conservative DeFi strategies, max 5% riskspawn an agent called research-bot to track AI agent news, update it after each sessionsave this thesis to vault

What you get

116 tools across four pillars:

Pillar	What it does
Memory	Full-text searchable memory + versioned vault + chronicle
Agents	Spawn, recall, update named agents — `agent_recall` also pulls related memory/vault context matching the agent's goal, not just its own logged updates
Workflows	Automations, monitors, packets, deep research (auto-saves reports + auto-links related past research)
Execution	Base DeFi, Robinhood Chain, market data, web, GitHub

Coding and research sessions persist the same way: deep_research auto-saves its report to vault and links it to related past reports; code_session_save does the same for coding/debugging sessions (vault_save type=code, versioned per project, auto-linked). Both exist so the next session — yours or another agent's — starts with real context instead of cold.

vault_save and agent_spawn also take an optional workspaceProject - the same named Projects a user organizes their Agents/vault content into on the webapp's Agents page. Pass a name and it's matched case-insensitively or created automatically (list_projects to browse what exists first). Hosted vault only - local-vault mode has no project concept.

Default palette is core (lighter context). Full set:

Fully local

Finch is the runtime. Your LLM is the brain. Your data stays yours.

bash

npx -y -p @finchagentic/[email protected] finch setup# enable local vault (and optional local memory)

npx -y -p @finchagentic/[email protected] finch setup# enable local vault (and optional local memory)

Scheduled/cloud features still need an account. Core memory, vault, and public-data tools work offline of Finch cloud.

Configuration

Variable	Purpose
`FINCH_SESSION_TOKEN`	Signed-in session (vault/memory/agents against your account)
`FINCH_API_KEY`	API key (`finch_sk_…`)
`FINCH_TOOLS`	`core` (default) · `all` · or palettes like `memory,defi`
`FINCH_PROVIDER`	Force `bankr` · `anthropic` · `openai` · `grok`
`FINCH_MODEL`	Model override for host-side loops
`BANKR_API_KEY` / `ANTHROPIC_API_KEY` / `OPENAI_API_KEY` / `GROK_API_KEY`	Required for `ask_finch`, `deep_research`'s synthesis stages, and scheduled-agent learning extraction — these do their own multi-step reasoning server-side and there is no free hosted fallback (BYOK only, one key is enough)
`FIRECRAWL_API_KEY`	Better crawl quality (optional)
`GITHUB_TOKEN`	For `github_search_code`
`ALCHEMY_API_KEY`	Faster Base RPC (optional)

Cost model: almost everything is free to run — the other 110 tools are plain API/RPC calls, and your MCP client's own model (Claude, GPT, whatever's driving the chat) does all the tool-selection reasoning at no cost to Finch. The 5 exceptions above need their own key because their reasoning happens inside the tool call, invisible to your client, and can't be delegated to it. Set exactly one of the four env vars and every tool that needs it will use it automatically.

Guided setup:

Security

#	Boundary	Rule
1	Prompt injection	External content is data only — never instructions
2	Mainnet confirm	Estimate → preview → confirm → execute
3	Pinned install	Always `@finchagentic/[email protected]`, never `@latest`
4	Credential vault	Never paste secrets into prompts or third-party tools
5	Data disclosure	Know what leaves the machine (LLM, Firecrawl, GitHub, chain RPCs)
6	Server monitors	Scheduled jobs need explicit confirmation
7	Fund-moving confirm	`stake_finch`/`unstake_finch`/`base_mcp_send`/`base_mcp_swap`/`rh_mcp_swap` all require `confirm: true`
8	Local wallet encryption	Set `FINCH_WALLET_PASSPHRASE` for a portable, passphrase-derived key — without it, the key derives from a random per-install secret + machine info, so the wallet file alone (without that secret file) isn't enough to decrypt it

Troubleshooting

Problem	Fix
Tools missing	Fully restart the MCP client
Old version	`npx clear-npx-cache` then restart
Auth issues	`finch login` or set `FINCH_API_KEY` / `FINCH_SESSION_TOKEN`
`ask_finch`/`deep_research` error: "No LLM provider configured"	Set one of `BANKR_API_KEY` / `ANTHROPIC_API_KEY` / `OPENAI_API_KEY` / `GROK_API_KEY` — see Configuration, no free fallback exists
Diagnose	`finch doctor`

Links

MIT License · Finch

來源：README.md，提交 14e0197

工具

SourceWeft 尚未讀取此伺服器的工具。

版本歷史

v4.7.0最新Sep 16, 2026

評分

已滿足適用檢查項的 60%

該評分衡量此收錄條目的完整程度，依據註冊中心條目、端點與公開來源計算得出。它不是對伺服器本身的評價，也不表示伺服器安全或目前可用：可連通性是帶日期的檢查結果，評分也不包含安裝量或使用者評價。

近 30 天內可連通必需25 分
請保持端點能夠回應 MCP 握手；SourceWeft 每 30 天探測一次。
已宣告認證方式必需10 分
已列出工具15 分
請讓伺服器在無憑證時回應 tools/list，或發布可在沙箱中無憑證啟動的軟體包。
描述必需15 分
圖示10 分
文件15 分
授權條款10 分

Piece	Location
Vault	`~/.finch/vault/`
Wallet	`~/.finch/wallet.json`
Config	`~/.finch/config.json`
Brain	your MCP client’s model