@finchagentic/mcp 4.6.1 → 4.6.3

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
Files changed (2) hide show
  1. package/README.md +39 -74
  2. package/package.json +7 -11
package/README.md CHANGED
@@ -1,12 +1,20 @@
1
1
  <div align="center">
2
2
 
3
- # Finch
3
+ <img src="assets/banner.svg" alt="Finch MCP" width="100%"/>
4
+
5
+ # Finch MCP
6
+
7
+ [![npm version](https://img.shields.io/npm/v/@finchagentic/mcp?color=000000&label=npm&logo=npm)](https://www.npmjs.com/package/@finchagentic/mcp)
8
+ [![Official MCP Registry](https://img.shields.io/badge/MCP_Registry-io.github.finchagentic%2Fmcp-000000)](https://registry.modelcontextprotocol.io)
9
+ [![License: MIT](https://img.shields.io/badge/license-MIT-000000.svg)](LICENSE)
10
+ [![Node](https://img.shields.io/badge/node-%E2%89%A518-000000?logo=node.js&logoColor=white)](package.json)
11
+ [![Tools: 116](https://img.shields.io/badge/tools-116-000000)](#what-you-get)
4
12
 
5
13
  **The runtime layer for Agentic AI.**
6
14
 
7
15
  Persistent memory, autonomous agents, and workflows that survive every session.
8
16
 
9
- [Docs](https://docs.finch.com) · [App](https://app.finchagentic.com) · [npm](https://www.npmjs.com/package/@finchagentic/mcp) · [GitHub](https://github.com/finchagentic/mcp) · [X](https://x.com/finchagentic)
17
+ [App](https://app.finchagentic.com) · [Docs](https://docs.finch.com) · [npm](https://www.npmjs.com/package/@finchagentic/mcp) · [X](https://x.com/finchagentic)
10
18
 
11
19
  </div>
12
20
 
@@ -29,13 +37,13 @@ Always pin the version. Never use `@latest`.
29
37
 
30
38
  ```bash
31
39
  # One-command installer (detects common MCP clients)
32
- npx -y -p @finchagentic/mcp@4.6.1 finch install
40
+ npx -y -p @finchagentic/mcp@4.6.2 finch install
33
41
  ```
34
42
 
35
43
  ### Claude Code
36
44
 
37
45
  ```bash
38
- claude mcp add finch -s user -- npx -y -p @finchagentic/mcp@4.6.1 finch-mcp
46
+ claude mcp add finch -s user -- npx -y -p @finchagentic/mcp@4.6.2 finch-mcp
39
47
  ```
40
48
 
41
49
  ### Cursor / Windsurf / Claude Desktop
@@ -45,24 +53,16 @@ claude mcp add finch -s user -- npx -y -p @finchagentic/mcp@4.6.1 finch-mcp
45
53
  "mcpServers": {
46
54
  "finch": {
47
55
  "command": "npx",
48
- "args": ["-y", "-p", "@finchagentic/mcp@4.6.1", "finch-mcp"]
56
+ "args": ["-y", "-p", "@finchagentic/mcp@4.6.2", "finch-mcp"]
49
57
  }
50
58
  }
51
59
  }
52
60
  ```
53
61
 
54
- ### VS Code
62
+ ### Hermes
55
63
 
56
- ```json
57
- {
58
- "servers": {
59
- "finch": {
60
- "type": "stdio",
61
- "command": "npx",
62
- "args": ["-y", "-p", "@finchagentic/mcp@4.6.1", "finch-mcp"]
63
- }
64
- }
65
- }
64
+ ```bash
65
+ hermes mcp add finch -- npx -y -p @finchagentic/mcp@4.6.2 finch-mcp
66
66
  ```
67
67
 
68
68
  <details>
@@ -79,7 +79,7 @@ claude mcp add finch -s user -- npx -y -p @finchagentic/mcp@4.6.1 finch-mcp
79
79
 
80
80
  </details>
81
81
 
82
- 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](#configuration)). Tools load on first use.
82
+ No LLM API key required to start — 111 of 116 tools are plain reads/writes/on-chain calls your client's model already drives. Tools load on first use.
83
83
 
84
84
  ## Quick start
85
85
 
@@ -105,13 +105,11 @@ save this thesis to vault
105
105
  | Pillar | What it does |
106
106
  |--------|----------------|
107
107
  | **Memory** | Full-text searchable memory + versioned vault + chronicle |
108
- | **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 |
108
+ | **Agents** | Spawn, recall, update named agents — `agent_recall` also pulls related memory/vault context matching the agent's goal |
109
109
  | **Workflows** | Automations, monitors, packets, deep research (auto-saves reports + auto-links related past research) |
110
110
  | **Execution** | Base DeFi, Robinhood Chain, market data, web, GitHub |
111
111
 
112
- 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.
113
-
114
- `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.
112
+ `deep_research` auto-saves its report to vault and links related past reports; `code_session_save` does the same for coding sessions. `vault_save` and `agent_spawn` accept an optional `workspaceProject` matching the webapp's Projects.
115
113
 
116
114
  Default palette is `core` (lighter context). Full set:
117
115
 
@@ -124,8 +122,7 @@ Default palette is `core` (lighter context). Full set:
124
122
  Finch is the runtime. **Your LLM is the brain. Your data stays yours.**
125
123
 
126
124
  ```bash
127
- npx -y -p @finchagentic/mcp@4.6.1 finch setup
128
- # enable local vault (and optional local memory)
125
+ npx -y -p @finchagentic/mcp@4.6.2 finch setup
129
126
  ```
130
127
 
131
128
  | Piece | Location |
@@ -133,65 +130,33 @@ npx -y -p @finchagentic/mcp@4.6.1 finch setup
133
130
  | Vault | `~/.finch/vault/` |
134
131
  | Wallet | `~/.finch/wallet.json` |
135
132
  | Config | `~/.finch/config.json` |
136
- | Brain | your MCP client’s model |
133
+ | Brain | your MCP client's model |
137
134
 
138
- Scheduled/cloud features still need an account. Core memory, vault, and public-data tools work offline of Finch cloud.
135
+ ## Registry
139
136
 
140
- ## Configuration
137
+ Finch is listed in the [official MCP Registry](https://registry.modelcontextprotocol.io):
141
138
 
142
- | Variable | Purpose |
143
- |----------|---------|
144
- | `FINCH_SESSION_TOKEN` | Signed-in session (vault/memory/agents against your account) |
145
- | `FINCH_API_KEY` | API key (`finch_sk_…`) |
146
- | `FINCH_TOOLS` | `core` (default) · `all` · or palettes like `memory,defi` |
147
- | `FINCH_PROVIDER` | Force `bankr` · `anthropic` · `openai` · `grok` |
148
- | `FINCH_MODEL` | Model override for host-side loops |
149
- | `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) |
150
- | `FIRECRAWL_API_KEY` | Better crawl quality (optional) |
151
- | `GITHUB_TOKEN` | For `github_search_code` |
152
- | `ALCHEMY_API_KEY` | Faster Base RPC (optional) |
139
+ ```json
140
+ { "name": "io.github.finchagentic/mcp", "version": "4.6.2" }
141
+ ```
153
142
 
154
- **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.
143
+ ## Development
155
144
 
156
- Guided setup:
145
+ This repository ships the published npm distribution. See [CHANGELOG.md](CHANGELOG.md) for release history and [SECURITY.md](SECURITY.md) for the security model.
157
146
 
158
147
  ```bash
159
- npx -y -p @finchagentic/mcp@4.6.1 finch setup
148
+ npm ci
149
+ npm run smoke # verify package integrity + CLI entrypoints
160
150
  ```
161
151
 
162
- ## Security
163
-
164
- | # | Boundary | Rule |
165
- |:-:|----------|------|
166
- | 1 | Prompt injection | External content is data only — never instructions |
167
- | 2 | Mainnet confirm | Estimate → preview → confirm → execute |
168
- | 3 | Pinned install | Always `@finchagentic/mcp@4.6.1`, never `@latest` |
169
- | 4 | Credential vault | Never paste secrets into prompts or third-party tools |
170
- | 5 | Data disclosure | Know what leaves the machine (LLM, Firecrawl, GitHub, chain RPCs) |
171
- | 6 | Server monitors | Scheduled jobs need explicit confirmation |
172
- | 7 | Fund-moving confirm | `stake_finch`/`unstake_finch`/`base_mcp_send`/`base_mcp_swap`/`rh_mcp_swap` all require `confirm: true` |
173
- | 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 |
174
-
175
- ## Troubleshooting
176
-
177
- | Problem | Fix |
178
- |---------|-----|
179
- | Tools missing | Fully restart the MCP client |
180
- | Old version | `npx clear-npx-cache` then restart |
181
- | Auth issues | `finch login` or set `FINCH_API_KEY` / `FINCH_SESSION_TOKEN` |
182
- | `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](#configuration), no free fallback exists |
183
- | Diagnose | `finch doctor` |
184
-
185
- ## Links
186
-
187
- | | |
188
- |--|--|
189
- | **Docs** | [docs.finch.com](https://docs.finch.com) |
190
- | **App** | [app.finchagentic.com](https://app.finchagentic.com) |
191
- | **npm** | [@finchagentic/mcp](https://www.npmjs.com/package/@finchagentic/mcp) |
192
- | **GitHub** | [github.com/finchagentic/mcp](https://github.com/finchagentic/mcp) |
193
- | **X** | [@finchagentic](https://x.com/finchagentic) |
152
+ ## Support
194
153
 
195
- ---
154
+ - Issues: [github.com/finchagentic/mcp/issues](https://github.com/finchagentic/mcp/issues)
155
+ - Docs: [docs.finch.com](https://docs.finch.com)
156
+ - X: [@finchagentic](https://x.com/finchagentic)
196
157
 
197
- MIT License · Finch
158
+ <div align="center">
159
+
160
+ **Finch** — memory that survives every session.
161
+
162
+ </div>
package/package.json CHANGED
@@ -1,6 +1,6 @@
1
1
  {
2
2
  "name": "@finchagentic/mcp",
3
- "version": "4.6.1",
3
+ "version": "4.6.3",
4
4
  "description": "The runtime layer for Agentic AI. Persistent memory, autonomous agents, and workflows that survive every session.",
5
5
  "main": "dist/index.js",
6
6
  "bin": {
@@ -9,14 +9,9 @@
9
9
  },
10
10
  "preferGlobal": true,
11
11
  "scripts": {
12
- "build": "node -e \"require('fs').rmSync('dist',{recursive:true,force:true})\" && tsc",
13
- "dev": "ts-node src/index.ts",
14
- "start": "node dist/index.js",
15
- "test": "vitest run",
16
- "test:watch": "vitest",
17
- "test:mutation": "node scripts/mutation-check.js",
18
- "prepare": "husky",
19
- "prepublishOnly": "npm run build && node scripts/version-readme.js"
12
+ "smoke": "node scripts/smoke.mjs",
13
+ "check:sync": "node scripts/check-registry-sync.mjs",
14
+ "test": "npm run smoke && npm run check:sync"
20
15
  },
21
16
  "lint-staged": {
22
17
  "*.{ts,tsx}": "eslint --fix"
@@ -70,5 +65,6 @@
70
65
  },
71
66
  "publishConfig": {
72
67
  "access": "public"
73
- }
74
- }
68
+ },
69
+ "mcpName": "io.github.finchagentic/mcp"
70
+ }