Security: add hardening module and secure-bot extension by AlphonseC · Pull Request #7346 · openclaw/openclaw

AlphonseC · 2026-02-02T18:32:22Z

Add src/security/hardening.ts with comprehensive security utilities:
- Cryptographically secure token generation
- Input validation and sanitization
- Path traversal prevention
- File permission hardening
- Rate limiting implementation
- Security headers helper
- Audit logging utilities
Add extensions/secure-bot with AI bot security features:
- Prompt injection detection with 20+ patterns
- Access control lists (allow/block/admin)
- Rate limiting per user
- Sensitive data redaction (PII, API keys)
- Security event logging and metrics
- Configurable security policies
Add comprehensive test suite for hardening module

https://claude.ai/code/session_0152DnGhhwvDXMppXwT6dtPz

Greptile Overview

Greptile Summary

Adds a new src/security/hardening.ts module with helpers for token generation, input/path validation, file permission hardening, config hardening/auditing, a simple in-memory rate limiter, and recommended HTTP security headers, along with a Vitest test suite.

Also introduces a new workspace extension package @openclaw/secure-bot implementing a channel plugin that applies access control, rate limiting, prompt-injection pattern detection, and redaction to inbound/outbound messages, plus basic schema/defaults for configuration.

Confidence Score: 2/5

This PR is not safe to merge as-is due to a runtime ESM/CJS incompatibility and likely plugin packaging issues.
The new sanitizePath uses require() inside an ESM TS module, which will throw at runtime when invoked. Separately, the @openclaw/secure-bot package is configured to publish dist/ but does not build or include compiled output, and plugin installs omit devDependencies, so the extension is likely non-functional when installed normally. Other issues are lower severity (duplicate exports, potentially broken docs link).
src/security/hardening.ts; extensions/secure-bot/package.json; extensions/secure-bot/src/index.ts

_{(2/5) Greptile learns from your feedback when you react with thumbs up/down!}

Context used:

Context from dashboard - CLAUDE.md (source)
Context from dashboard - AGENTS.md (source)

- Add src/security/hardening.ts with comprehensive security utilities: - Cryptographically secure token generation - Input validation and sanitization - Path traversal prevention - File permission hardening - Rate limiting implementation - Security headers helper - Audit logging utilities - Add extensions/secure-bot with AI bot security features: - Prompt injection detection with 20+ patterns - Access control lists (allow/block/admin) - Rate limiting per user - Sensitive data redaction (PII, API keys) - Security event logging and metrics - Configurable security policies - Add comprehensive test suite for hardening module https://claude.ai/code/session_0152DnGhhwvDXMppXwT6dtPz

greptile-apps

_{3 files reviewed, 4 comments}

_{Edit Code Review Agent Settings | Greptile}

greptile-apps · 2026-02-02T18:34:54Z

src/security/hardening.ts

+export function sanitizePath(inputPath: string, baseDir: string): string | null {
+  const path = require("node:path");
+


[P0] sanitizePath uses CommonJS require() in an ESM TypeScript module.

In this repo, src/**/*.ts is ESM (import ... from "node:path"), so require("node:path") will throw at runtime in ESM contexts. This will break any code path that calls sanitizePath.

Suggested change

export function sanitizePath(inputPath: string, baseDir: string): string | null {

const path = require("node:path");

import * as path from "node:path";

Prompt To Fix With AI

This is a comment left during a code review. Path: src/security/hardening.ts Line: 226:228 Comment: [P0] `sanitizePath` uses CommonJS `require()` in an ESM TypeScript module. In this repo, `src/**/*.ts` is ESM (`import ... from "node:path"`), so `require("node:path")` will throw at runtime in ESM contexts. This will break any code path that calls `sanitizePath`. ```suggestion import * as path from "node:path"; ``` How can I resolve this? If you propose a fix, please make it concise.

greptile-apps · 2026-02-02T18:34:55Z

extensions/secure-bot/package.json

+  "license": "MIT",
+  "peerDependencies": {
+    "openclaw": "*"
+  },
+  "devDependencies": {
+    "openclaw": "workspace:*",
+    "typescript": "^5.3.0",
+    "vitest": "^1.0.0"
+  },
+  "files": [


[P0] Plugin install will omit devDependencies, but typescript/vitest are only in devDependencies and there is no dist/ committed.

Per repo guidelines, plugin installs run npm install --omit=dev in the extension dir. With the current files: ["dist", ...] and no build-on-install script, a published package will ship without compiled JS and can't run. Either ship prebuilt dist in the published tarball or move any runtime-required deps/scripts accordingly.

Prompt To Fix With AI

This is a comment left during a code review. Path: extensions/secure-bot/package.json Line: 27:36 Comment: [P0] Plugin install will omit `devDependencies`, but `typescript`/`vitest` are only in `devDependencies` and there is no `dist/` committed. Per repo guidelines, plugin installs run `npm install --omit=dev` in the extension dir. With the current `files: ["dist", ...]` and no build-on-install script, a published package will ship without compiled JS and can't run. Either ship prebuilt `dist` in the published tarball or move any runtime-required deps/scripts accordingly. How can I resolve this? If you propose a fix, please make it concise.

greptile-apps · 2026-02-02T18:34:56Z

extensions/secure-bot/src/index.ts

+// Export security engine for direct access
+export { SecurityEngine, securityEngine, DEFAULT_CONFIG };
+
+// Export types
+export type { SecureBotConfig, SecurityEvent, SecurityEventType, SecurityMetrics };
+
+export default plugin;


[P1] Duplicate exports for the same types.

SecureBotConfig, SecurityEvent, SecurityEventType, and SecurityMetrics are already exported as export type ... where they are declared, then re-exported again at the bottom. This tends to trip linting/TS settings (duplicate export) and makes the public surface harder to reason about.

Also appears in the same file for other already-exported items: DEFAULT_CONFIG is exported twice (const + re-export).

Prompt To Fix With AI

This is a comment left during a code review. Path: extensions/secure-bot/src/index.ts Line: 669:675 Comment: [P1] Duplicate exports for the same types. `SecureBotConfig`, `SecurityEvent`, `SecurityEventType`, and `SecurityMetrics` are already exported as `export type ...` where they are declared, then re-exported again at the bottom. This tends to trip linting/TS settings (duplicate export) and makes the public surface harder to reason about. Also appears in the same file for other already-exported items: `DEFAULT_CONFIG` is exported twice (const + re-export). How can I resolve this? If you propose a fix, please make it concise.

greptile-apps · 2026-02-02T18:34:57Z

extensions/secure-bot/src/index.ts

+const meta: ChannelMeta = {
+  id: "secure-bot",
+  label: "Secure Bot",
+  description: "Security-hardened AI bot with enhanced protection features",
+  emoji: "🛡️",
+  color: "#4CAF50",
+  docs: "https://docs.openclaw.ai/extensions/secure-bot",
+};


[P1] Hardcoded docs URL likely wrong.

meta.docs points to https://docs.openclaw.ai/extensions/secure-bot, but the README and repo don't add the corresponding docs page in this PR. If this page doesn't exist, the in-product docs link will be broken for users.

Prompt To Fix With AI

This is a comment left during a code review. Path: extensions/secure-bot/src/index.ts Line: 555:562 Comment: [P1] Hardcoded docs URL likely wrong. `meta.docs` points to `https://docs.openclaw.ai/extensions/secure-bot`, but the README and repo don't add the corresponding docs page in this PR. If this page doesn't exist, the in-product docs link will be broken for users. How can I resolve this? If you propose a fix, please make it concise.

新增資安漏洞修復模組，涵蓋 Top 10 漏洞： 1. Gateway 暴露於 0.0.0.0:18789 - 自動產生安全 token - 強制 loopback 綁定 2. DM policy 允許所有使用者 - 偵測 open policy - 自動切換為 allowlist 3. Sandbox 預設停用 - 啟用完整沙箱模式 - Docker 網路隔離 4. Credentials 明文儲存 - 強制 600/700 權限 - 目錄安全檢查 5. Prompt injection 透過 web content - 20+ 注入模式偵測 - 自動包裝不信任內容 6. 危險命令未封鎖 - rm -rf, curl|sh 等封鎖清單 - 命令安全檢查函式 7. 無網路隔離 - Docker network=none - 移除 DNS/extraHosts 8. 過高工具存取權限 - 最小權限建議 - 工具白名單 9. 無稽核日誌 - 啟用日誌遮蔽 - Session 記錄建議 10. 弱配對碼 - 12 字元安全碼 - 移除易混淆字元 https://claude.ai/code/session_0152DnGhhwvDXMppXwT6dtPz

新增 Python 依賴檔案，用於： - 整合 OpenClaw Gateway API - 自動化安全稽核腳本 - WebSocket 客戶端連接 https://claude.ai/code/session_0152DnGhhwvDXMppXwT6dtPz

openclaw-barnacle · 2026-02-28T15:12:41Z

Please make this as a third-party plugin that you maintain yourself in your own repo. Docs: https://docs.openclaw.ai/plugin. Feel free to open a PR after to add it to our community plugins page: https://docs.openclaw.ai/plugins/community

greptile-apps bot reviewed Feb 2, 2026

View reviewed changes

Reapor-Yurnero mentioned this pull request Feb 3, 2026

feat(gateway): support modular guardrails extensions for securing against indirect prompt injections and other agentic threats #6095

Closed

Build: add Python requirements.txt for integration tools

7d510f4

新增 Python 依賴檔案，用於： - 整合 OpenClaw Gateway API - 自動化安全稽核腳本 - WebSocket 客戶端連接 https://claude.ai/code/session_0152DnGhhwvDXMppXwT6dtPz

thewilloftheshadow force-pushed the main branch from bfc1ccb to f92900f Compare February 15, 2026 18:46

thewilloftheshadow added the r: third-party-extension label Feb 28, 2026

openclaw-barnacle bot closed this Feb 28, 2026

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Uh oh!

Security: add hardening module and secure-bot extension#7346

Security: add hardening module and secure-bot extension#7346
AlphonseC wants to merge 3 commits intoopenclaw:mainfrom
AlphonseC:claude/security-analysis-ai-bot-NdOjG

AlphonseC commented Feb 2, 2026 •

edited by greptile-apps bot

Loading

Uh oh!

greptile-apps bot left a comment

Uh oh!

greptile-apps bot Feb 2, 2026

Uh oh!

greptile-apps bot Feb 2, 2026

Uh oh!

greptile-apps bot Feb 2, 2026

Uh oh!

greptile-apps bot Feb 2, 2026

Uh oh!

openclaw-barnacle bot commented Feb 28, 2026

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants

		export function sanitizePath(inputPath: string, baseDir: string): string \| null {
		const path = require("node:path");

	export function sanitizePath(inputPath: string, baseDir: string): string \| null {
	const path = require("node:path");
	import * as path from "node:path";

Uh oh!

Conversation

AlphonseC commented Feb 2, 2026 • edited by greptile-apps bot Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Greptile Overview

Greptile Summary

Confidence Score: 2/5

Uh oh!

greptile-apps bot left a comment

Choose a reason for hiding this comment

Uh oh!

greptile-apps bot Feb 2, 2026

Choose a reason for hiding this comment

Uh oh!

greptile-apps bot Feb 2, 2026

Choose a reason for hiding this comment

Uh oh!

greptile-apps bot Feb 2, 2026

Choose a reason for hiding this comment

Uh oh!

greptile-apps bot Feb 2, 2026

Choose a reason for hiding this comment

Uh oh!

openclaw-barnacle bot commented Feb 28, 2026

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants

AlphonseC commented Feb 2, 2026 •

edited by greptile-apps bot

Loading