--- name: safety-guard description: Safety Guard URLs or files with the safety-guard CLI (web, PDFs, images, audio, YouTube). homepage: https://safety-guard.sh metadata: {"clawdbot":{"emoji":"🧾","r
Runtime security guard + scanner for OpenClaw agents. Part of the guard-scanner ecosystem. Detects reverse shells, credential theft, and sandbox escapes in r...
JEP Guard intercepts high-risk commands, requires user confirmation, issues temporary tokens, and logs actions with exportable audit receipts.
--- name: content-safety-guard description: Dual-layer AI content guardrail with red-team test methodology metadata: {"openclaw": {"emoji": "🛡️", "os": ["darwin", "linux"], "requires": {"env": ["
Create an image of a Latino private security guard. The guard should be depicted wearing a tactical helmet and a bulletproof vest. The vest should have a communication radio attached and prominently d
Security guard skill for OpenClaw - Analyzes user input for harmful content, risky commands, and security threats before invoking LLM
--- name: oc_guard description: Safe OpenClaw config planning/apply workflow with bilingual execution receipts. metadata: {"openclaw":{"requires":{"bins":["python3","openclaw","opencode"]}}} --- # oc
--- name: wip-file-guard description: Hook that blocks destructive edits to protected identity files. For Claude Code CLI and OpenClaw. license: MIT interface: [cli, module, hook, plugin, skill] metad
Guard skill releases with pre-publish validation, quality checks, and safety gates. Use before publishing any skill to ensure it meets minimum standards, pas...
Security scanner and runtime guard for AI agent skills. 358 static threat patterns across 35 categories + 27 runtime checks (5 defense layers). Use when scan...
授权保护技能 - 所有外部 API 操作必须经过用户明确授权。这是最高优先级的安全层，确保没有任何自动化可以擅自使用你的授权。核心原则：用户指令
Policy-gated execution + tamper-evident audit trail for VAIBot/OpenClaw operations. Use to precheck/deny/require-approval before shell execution, and to prod...
Provide AI-powered security analysis and safety instruction tools to protect AI agents during MCP interactions. Analyze text content for harmful or inappropriate material and enhance user prompts with
Persona-weighted governance for outbound publishing (blog, social, announcements). Prevents unsafe public claims via hard-block checks, weighted consensus, r...
Monitors openclaw.json for changes, backs up before edits, restarts gateway, and auto-rolls back if no message received within a configurable timeout.
Persona-weighted merge governance for AI-assisted engineering. Evaluates PR risk (tests, security markers, reliability signals), returns MERGE/BLOCK/REVISE d...
Persona-weighted pre-send email governance for AI systems. Produces APPROVE/BLOCK/REWRITE decisions, writes decision artifacts to the board ledger, and retur...
Risk-aware support response governance with persona-weighted consensus. Detects legal/sensitive/confidentiality issues, applies hard-block policy checks, and...
Scan untrusted external text (web pages, tweets, search results, API responses) for prompt injection attacks. Returns severity levels and alerts on dangerous content. Use BEFORE processing any text fr
Pre-execution governance for high-risk agent actions. Uses persona-weighted consensus to decide ALLOW/BLOCK/REQUIRE_REWRITE before external or irreversible s...
License compliance for your own repos. Ensures correct copyright headers, dual-license blocks, and LICENSE files across all source files.
Open-source Consensus.Tools skill for governed AI decisions with board-native artifacts, strict JSON contracts, and deterministic policy behavior.
Monitors and verifies agent workspace files to detect unauthorized changes, injection attacks, personality drift, and cross-agent contamination.