@solongate/proxy 0.47.3 → 0.47.4

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 +5 -60
  2. package/package.json +1 -1
package/README.md CHANGED
@@ -1,5 +1,10 @@
1
1
  # @solongate/proxy
2
2
 
3
+ > **Don't `npm i` this.** There's nothing to install. Run the one-liner below to pair your machine:
4
+ > ```bash
5
+ > npx -y @solongate/proxy@latest login
6
+ > ```
7
+
3
8
  **Security for AI agents.** SolonGate enforces a security policy on every action an AI agent takes — every tool call is allowed, blocked, or logged before it runs. No code changes.
4
9
 
5
10
  ```
@@ -52,66 +57,6 @@ npx @solongate/proxy init --global --restore
52
57
 
53
58
  ---
54
59
 
55
- ## Advanced: wrap a single MCP server (proxy mode)
56
-
57
- Instead of the machine-wide guard, you can place SolonGate **in front of one MCP server** so it works with any MCP client (Claude Desktop, Cursor, Windsurf, Cline, Zed, …). Edit your `.mcp.json`:
58
-
59
- ```json
60
- {
61
- "mcpServers": {
62
- "my-server": {
63
- "command": "npx",
64
- "args": ["@solongate/proxy", "--policy", "restricted", "--", "node", "./my-server/dist/index.js"]
65
- }
66
- }
67
- }
68
- ```
69
-
70
- ### Policy presets
71
-
72
- | Preset | Description |
73
- |--------|-------------|
74
- | `restricted` | Blocks shell/exec/eval, allows reads and writes **(recommended)** |
75
- | `read-only` | Only allows read/list/get/search/query operations |
76
- | `permissive` | Allows all tools, still enforces input validation |
77
- | `deny-all` | Blocks all tool calls (emergency lockdown) |
78
-
79
- ### Custom policy file
80
-
81
- ```json
82
- {
83
- "id": "my-policy",
84
- "name": "My Custom Policy",
85
- "version": 1,
86
- "rules": [
87
- { "id": "deny-exec", "effect": "DENY", "priority": 100, "toolPattern": "*exec*", "permission": "EXECUTE", "minimumTrustLevel": "UNTRUSTED", "enabled": true },
88
- { "id": "allow-rest", "effect": "ALLOW", "priority": 1000, "toolPattern": "*", "permission": "EXECUTE", "minimumTrustLevel": "UNTRUSTED", "enabled": true }
89
- ]
90
- }
91
- ```
92
-
93
- ```bash
94
- npx @solongate/proxy --policy ./my-policy.json -- node my-server.js
95
- ```
96
-
97
- ### CLI options
98
-
99
- ```
100
- solongate-proxy [options] -- <server-command> [args...]
101
-
102
- --policy <preset|file> Policy preset or JSON file (default: restricted)
103
- --name <name> Proxy display name
104
- --verbose Show detailed error messages
105
- --no-input-guard Disable input validation
106
- --rate-limit <n> Per-tool rate limit (calls/min)
107
- --global-rate-limit <n> Global rate limit (calls/min)
108
- --config <file> Load full config from JSON file
109
- --api-key <key> SolonGate Cloud API key (cloud policy + audit)
110
- --api-url <url> Custom API URL (default: api.solongate.com)
111
- ```
112
-
113
- ---
114
-
115
60
  ## Why SolonGate?
116
61
 
117
62
  AI agents get direct access to your system — shell, file system, databases, network. A single prompt-injection attack can turn your assistant into an attacker. SolonGate sits between the agent and the action, enforcing your policy on every tool call before it runs, and recording every decision for audit.
package/package.json CHANGED
@@ -1,6 +1,6 @@
1
1
  {
2
2
  "name": "@solongate/proxy",
3
- "version": "0.47.3",
3
+ "version": "0.47.4",
4
4
  "description": "AI tool security proxy — protect any AI tool server with customizable policies, path/command constraints, rate limiting, and audit logging. Zero code changes required.",
5
5
  "type": "module",
6
6
  "bin": {