@poteshniy/agenttrust-mcp 1.0.0 → 1.0.1
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.
- package/README.md +46 -0
- package/package.json +3 -3
package/README.md
ADDED
|
@@ -0,0 +1,46 @@
|
|
|
1
|
+
# AgentTrust MCP Server
|
|
2
|
+
|
|
3
|
+
Security scanner and reputation oracle for AI agent skills and MCP servers.
|
|
4
|
+
|
|
5
|
+
## Install
|
|
6
|
+
|
|
7
|
+
Add to your MCP client config:
|
|
8
|
+
|
|
9
|
+
```json
|
|
10
|
+
{
|
|
11
|
+
"mcpServers": {
|
|
12
|
+
"agenttrust": {
|
|
13
|
+
"command": "npx",
|
|
14
|
+
"args": ["-y", "@poteshniy/agenttrust-mcp"]
|
|
15
|
+
}
|
|
16
|
+
}
|
|
17
|
+
}
|
|
18
|
+
```
|
|
19
|
+
|
|
20
|
+
**Claude Desktop:** `~/Library/Application Support/Claude/claude_desktop_config.json`
|
|
21
|
+
**Cursor:** `~/.cursor/mcp.json`
|
|
22
|
+
**Windsurf:** `~/.codeium/windsurf/mcp_config.json`
|
|
23
|
+
|
|
24
|
+
## Tools
|
|
25
|
+
|
|
26
|
+
| Tool | Price | Description |
|
|
27
|
+
|---|---|---|
|
|
28
|
+
| `scan_skill_free` | FREE | SKILL.md scan, 5 rules |
|
|
29
|
+
| `scan_skill` | $0.015 USDC | Full SKILL.md scan, 40 rules |
|
|
30
|
+
| `scan_mcp_free` | FREE | MCP manifest scan, 3 rules |
|
|
31
|
+
| `scan_mcp` | $0.015 USDC | Full MCP scan, 50 rules |
|
|
32
|
+
| `trust_gate` | FREE | Unified ACT/HALT gate |
|
|
33
|
+
| `check_reputation` | FREE | x402 endpoint reputation |
|
|
34
|
+
| `verify_hash` | $0.005 USDC | Verify skill hash |
|
|
35
|
+
| `full_report` | $0.050 USDC | Full audit report |
|
|
36
|
+
| `wallet_reputation` | $0.010 USDC | Wallet trust score |
|
|
37
|
+
| `get_badge_url` | FREE | Get SVG trust badge URL |
|
|
38
|
+
|
|
39
|
+
## Example
|
|
40
|
+
|
|
41
|
+
Ask your agent:
|
|
42
|
+
> "Scan this SKILL.md before installing"
|
|
43
|
+
> "Check if this MCP server is safe to connect"
|
|
44
|
+
> "What's the trust score for this endpoint?"
|
|
45
|
+
|
|
46
|
+
Powered by [AgentTrust](https://agenttrust.uk) — x402-native, no API keys required.
|
package/package.json
CHANGED
|
@@ -1,7 +1,7 @@
|
|
|
1
1
|
{
|
|
2
2
|
"name": "@poteshniy/agenttrust-mcp",
|
|
3
|
-
"version": "1.0.
|
|
4
|
-
"description": "AgentTrust MCP Server
|
|
3
|
+
"version": "1.0.1",
|
|
4
|
+
"description": "AgentTrust MCP Server — security scanner for AI agent skills and MCP servers",
|
|
5
5
|
"main": "index.js",
|
|
6
6
|
"bin": {
|
|
7
7
|
"agenttrust-mcp": "./index.js"
|
|
@@ -27,4 +27,4 @@
|
|
|
27
27
|
"type": "git",
|
|
28
28
|
"url": "https://github.com/poteshniy/agenttrust"
|
|
29
29
|
}
|
|
30
|
-
}
|
|
30
|
+
}
|