mcp-sentinel 0.2.1 → 0.2.2

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 CHANGED
@@ -12,7 +12,7 @@
12
12
  </p>
13
13
  </p>
14
14
 
15
- > **v0.2.0** -- Risk scoring (A-F), SARIF output, per-tool security analysis, verbose mode, HTTP header support
15
+ > **v0.2.1** -- Security hardening, Unicode evasion defense, parameter-aware categorization, ReDoS prevention
16
16
 
17
17
  ---
18
18
 
@@ -29,12 +29,14 @@ MCP Sentinel connects to any MCP server, shows you every tool it exposes, assign
29
29
  - **Risk Scoring** -- A-F grade for every server based on tool risk, security findings, and attack surface
30
30
  - **Policy Engine** -- YAML-based deny/require/allow rules with glob patterns and auto-detection
31
31
  - **Deep Security Analysis** -- Per-tool scanning with [Aguara](https://github.com/garagon/aguara) (177 rules: prompt injection, exfiltration, credential leaks)
32
- - **Smart Categorization** -- Tools auto-escalate from "read" to "admin" when critical findings are detected
32
+ - **Smart Categorization** -- Analyzes tool names, descriptions, and parameters; auto-escalates when critical findings are detected
33
33
  - **Multi-Transport** -- stdio, SSE, and Streamable HTTP with custom header support
34
34
  - **Config Discovery** -- Auto-scan servers from Claude Desktop, Cursor, Windsurf, VS Code, Zed
35
35
  - **CI/CD Ready** -- SARIF output for GitHub Code Scanning, exit codes for policy violations
36
36
  - **Drift Detection** -- Save baselines and detect added/removed/changed tools over time
37
37
  - **Multiple Exports** -- Terminal, JSON, Markdown, SARIF
38
+ - **Unicode Evasion Defense** -- NFKC normalization prevents homoglyph and fullwidth character bypasses
39
+ - **Hardened Inputs** -- Path traversal prevention, header injection blocking, ReDoS-safe policy patterns
38
40
 
39
41
  ## Quick Start
40
42
 
@@ -47,7 +49,7 @@ That's it. You'll see every tool the server exposes, categorized by risk:
47
49
 
48
50
  ```
49
51
  ┌──────────────────────────────┐
50
- │ MCP Sentinel v0.2.0
52
+ │ MCP Sentinel v0.2.1
51
53
  └──────────────────────────────┘
52
54
 
53
55
  Server secure-filesystem-server v0.2.0
package/dist/version.d.ts CHANGED
@@ -1,2 +1,2 @@
1
- export declare const VERSION = "0.2.1";
1
+ export declare const VERSION = "0.2.2";
2
2
  //# sourceMappingURL=version.d.ts.map
package/dist/version.js CHANGED
@@ -1,2 +1,2 @@
1
- export const VERSION = "0.2.1";
1
+ export const VERSION = "0.2.2";
2
2
  //# sourceMappingURL=version.js.map
package/package.json CHANGED
@@ -1,6 +1,6 @@
1
1
  {
2
2
  "name": "mcp-sentinel",
3
- "version": "0.2.1",
3
+ "version": "0.2.2",
4
4
  "description": "Scan, score, and enforce security policies on MCP servers. Risk scoring (A-F), SARIF output, policy engine, drift detection, aguara integration, CI/CD gate.",
5
5
  "author": "Gustavo Aragon (@oktsec)",
6
6
  "license": "Apache-2.0",