Plugins
OpenClaw sandbox backend for the NVIDIA OpenShell CLI with mirrored local workspaces and SSH command execution.
AI agent security plugin for OpenClaw: prompt injection detection, PII sanitization, and monitoring dashboard
Skill-vetter preflight reports and install-path enforcement for OpenClaw skills/plugins.
OpenClaw plugin: route model calls through the local cursor-agent binary (Cursor subscription). Adds a cursor-cli provider and CLI backend with live model-catalog refresh and unified --thinking mapping.
Connects to AxonFlow (SaaS or self-hosted) for policy enforcement, PII scanning, approval gates, and audit trails in OpenClaw agents
OpenClaw security guard plugin — tool-call interception, approval flows, and safety rules
Protect API keys and secrets for OpenClaw. Credentials stay in your vault, never in the agent's memory
OpenClaw plugin exposing IdentyClaw API agent tools
OpenClaw plugin — permission and approval engine. Gates built-in tool calls (bash, file edit, etc.) and plugin tools via a three-bucket policy (allow/deny/ask) with wildcard rules, rule sources (session/workspace/user/config), and in-chat approval. No network calls.
OpenClaw security plugin — runtime tool call interception, command-level allow/deny/approve, bypass detection, file/network path rules, and audit logging
Universal security-audit plugin and CLI engine for AI agents with OWASP-mapped reports.
Automatic prompt-injection screening for OpenCLAW tool calls. Auto-registers a before_tool_call hook that screens web fetch/search results and other untrusted tool content.
OpenClaw plugin for privacy-preserving LLM interactions via PII redaction
Safe OpenClaw auth and model-state reset plugin by Growthcircle.id.
Security enforcement layer for Claude Code. Blocks dangerous commands, audits every tool call, detects prompt injection.
Stop your AI agent from leaking private data or obeying prompt injections on a WhatsApp line: a senior AI reviews every outbound message and holds it until it's deliberately safe to send. AI-in-the-loop audit gate / guardrail for outbound channels.
Ed25519-signed, SHA-256-chained tool-call audit log for OpenClaw multi-agent setups. Tamper-evident provenance chain.
Gate OpenClaw tool calls against live Kinde roles, permissions, and feature flags.
Lumin observability + Agent Firewall for OpenClaw — captures prompts, responses, tool I/O, and synchronously enforces firewall policies on every tool call.
Guardrail plugin for OpenClaw that blocks prompt injection, jailbreak attempts, and sensitive credential exfiltration before agent dispatch.
Native OpenClaw plugin — secrets broker, MCP gateway enforcement, and approval flow for OpenClaw agents. Install via openclaw plugins install npm:@secr/openclaw-plugin (or clawhub:secr once published).
Pre-install security audit and vulnerability scanner for ClawHub skills and code plugins — scan by slug or local path, 9 threat intel sources, 7 checks including malware scanning, dependency vulnerabilities, and credential leak detection. Wedge paid API available at https://node1.tail507e53.ts.net/frisk for automated scanning at scale.