@agnt-id/mcp 0.1.0 → 0.1.1

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 +103 -0
  2. package/package.json +13 -3
package/README.md ADDED
@@ -0,0 +1,103 @@
1
+ # @agnt-id/mcp
2
+
3
+ MCP tool server for resolving `.agnt` names — works with Claude, Cursor, Windsurf, and any MCP-compatible client.
4
+
5
+ Exposes 6 tools that let AI agents resolve `.agnt` names, look up wallets, discover other agents, and check trust scores — all through the [Model Context Protocol](https://modelcontextprotocol.io).
6
+
7
+ ## Quick Start
8
+
9
+ No install required — run directly with npx:
10
+
11
+ ```bash
12
+ npx @agnt-id/mcp
13
+ ```
14
+
15
+ ### Claude Desktop
16
+
17
+ Add to your Claude Desktop config (`~/Library/Application Support/Claude/claude_desktop_config.json`):
18
+
19
+ ```json
20
+ {
21
+ "mcpServers": {
22
+ "agnt": {
23
+ "command": "npx",
24
+ "args": ["@agnt-id/mcp"]
25
+ }
26
+ }
27
+ }
28
+ ```
29
+
30
+ ### Cursor
31
+
32
+ Add to `.cursor/mcp.json` in your project:
33
+
34
+ ```json
35
+ {
36
+ "mcpServers": {
37
+ "agnt": {
38
+ "command": "npx",
39
+ "args": ["@agnt-id/mcp"]
40
+ }
41
+ }
42
+ }
43
+ ```
44
+
45
+ ### Claude Code
46
+
47
+ ```bash
48
+ claude mcp add agnt -- npx @agnt-id/mcp
49
+ ```
50
+
51
+ ## Tools
52
+
53
+ | Tool | Description |
54
+ |------|-------------|
55
+ | `resolve_agnt` | Resolve a `.agnt` name to wallets, endpoints, and metadata |
56
+ | `reverse_agnt` | Reverse lookup — find the `.agnt` name for a wallet address |
57
+ | `search_agnt` | Discover agents by keyword, skill, or domain |
58
+ | `check_availability` | Check if a `.agnt` name is available for registration |
59
+ | `owner_names` | List all `.agnt` names owned by an address |
60
+ | `trust_score` | Get trust score and signals for a `.agnt` name |
61
+
62
+ ### Example Prompts
63
+
64
+ Once connected, try asking your AI assistant:
65
+
66
+ - *"Resolve alice.agnt"*
67
+ - *"What agents have MCP endpoints?"*
68
+ - *"Is the name 'trading' available on agnt.id?"*
69
+ - *"Look up the trust score for oracle.agnt"*
70
+ - *"Who owns 0x1234...abcd on agnt.id?"*
71
+
72
+ ## Environment Variables
73
+
74
+ | Variable | Default | Description |
75
+ |----------|---------|-------------|
76
+ | `AGNT_GATEWAY` | `https://api.agnt.id` | Gateway API base URL |
77
+ | `AGNT_CHAIN` | `eip155:1` | CAIP-2 chain ID (1 = Ethereum, 8453 = Base, 42161 = Arbitrum) |
78
+
79
+ Example with custom gateway:
80
+
81
+ ```json
82
+ {
83
+ "mcpServers": {
84
+ "agnt": {
85
+ "command": "npx",
86
+ "args": ["@agnt-id/mcp"],
87
+ "env": {
88
+ "AGNT_CHAIN": "eip155:8453"
89
+ }
90
+ }
91
+ }
92
+ }
93
+ ```
94
+
95
+ ## Links
96
+
97
+ - [Website](https://www.agnt.id)
98
+ - [API Docs](https://www.agnt.id/docs)
99
+ - [GitHub](https://github.com/txc0ld/agnt.id)
100
+
101
+ ## License
102
+
103
+ MIT
package/package.json CHANGED
@@ -1,6 +1,6 @@
1
1
  {
2
2
  "name": "@agnt-id/mcp",
3
- "version": "0.1.0",
3
+ "version": "0.1.1",
4
4
  "description": "MCP tool server for resolving .agnt names",
5
5
  "type": "module",
6
6
  "bin": {
@@ -8,13 +8,23 @@
8
8
  },
9
9
  "main": "dist/server.js",
10
10
  "types": "dist/server.d.ts",
11
- "files": ["dist", "README.md"],
11
+ "files": [
12
+ "dist",
13
+ "README.md"
14
+ ],
12
15
  "scripts": {
13
16
  "build": "tsc",
14
17
  "start": "node dist/server.js",
15
18
  "prepublishOnly": "tsc"
16
19
  },
17
- "keywords": ["agnt", "mcp", "agent", "naming", "resolve", "model-context-protocol"],
20
+ "keywords": [
21
+ "agnt",
22
+ "mcp",
23
+ "agent",
24
+ "naming",
25
+ "resolve",
26
+ "model-context-protocol"
27
+ ],
18
28
  "license": "MIT",
19
29
  "dependencies": {
20
30
  "@modelcontextprotocol/sdk": "^1.12.1",