ClawGuard
Security tool for OpenClaw, covering dangerous action approval, scanning skills, blocking secret leaks.
GraphCanon updated Sep 12, 2026 · GitHub synced Sep 12, 2026
21views this month
Decision brief
ClawGuard is an AI security agent for OpenClaw that focuses on enhancing safety through the approval of risky actions, scanning skills and secrets for security issues, maintaining audit logs for critical processes.
Good fit when
- If you are using the OpenClaw environment and require specific security measures tailored to this ecosystem.
- When you need a tool that integrates seamlessly with runtime and data loss prevention techniques in an AI application context.
Avoid when
- Do not use ClawGuard if your project is not within the OpenClaw framework, as its features are designed for integration with it.
- Avoid using this tool if you require a plugin that has been formally released or certified; ClawGuard is still in development and uses metadata-only packages, without formal releases.
- Requirements:
- Installation of ClawGuard requires the use of PowerShell commands or pnpm for package management. This is specific to its context within OpenClaw.
Observed Jul 16, 2026 · Source: enrich:decision_facts
Verify the decision
Adoption
Package downloads where a registry match exists. GitHub stars (87) are secondary evidence.
- npm downloads (30d)
- 41·npm downloads API·Sep 12, 2026
Maintenance and security
Full trust report- Maintenance
- Slowing (175d since push)
- As of Sep 12, 2026
- Provenance
- Not a fork · Personal account
- As of Sep 12, 2026
- Security (OSV)
- 10 low (10 low)
- As of Jul 15, 2026
Public GitHub metadata and optional OSV scans. Signals, not a guarantee. Trust methodology.
Install
npm install ClawGuardSimilar tools
Same-category neighbours. No typed graph edges are catalogued for this tool yet.
Evidence and technical details
Sourced facts, taxonomy, compatibility claims, README excerpt, and machine-readable endpoints.
Overview
ClawGuard is an AI security agent designed to enhance safety within the OpenClaw environment by approving risky actions, scanning skills and secrets for security issues, and maintaining human oversight in critical processes. It operates as a plugin for OpenClaw, emphasizing runtime and data loss prevention, prompt injection defense, and audit logging.
Capability facts
- MCP server
- No MCP server detected
Source: repo_scan · Sep 12, 2026
- Languages
- typescript, javascript
Source: github.language+package.json · Sep 12, 2026
Categories
Tags
README
Install demo The install demo entry lives here: Recommended install method from the repo root: Optional local tarball demo only: Important posture: this is an install demo only , it is not published to a registry , is currently metadata only and still unpublished , and this READM...
For agents
This page has a .md twin and JSON over the API.