Destructive Command Guard - High-performance Rust hook for Claude Code that blocks dangerous commands before execution. SIMD-accelerated, modular pack system...

Switch OpenClaw heartbeat model to local Ollama and deploy a guard to prevent unauthorized changes, avoiding paid cloud token usage.

Lightweight two-stage input safety guard for agents. Use this skill when an agent must screen user input before answering, block prompt injection or prompt l...

JEP Guard intercepts high-risk commands, requires user confirmation, issues temporary tokens, and logs actions with exportable audit receipts.

OpenClaw Gateway Resilience Guard keeps Gateway, channels, logs, and optional model-provider probes observable after Wi-Fi changes, WSL sleep/resume, session...

Safety Guard URLs or files with the safety-guard CLI (web, PDFs, images, audio, YouTube).

Publish Guard is a public ClawHub pre-release audit skill. Use it when the user says "publish guard", "release audit", "pre-release check", or wants to revie...

Quiet logic guard for risky multi-skill workflows. Checks order before delete, send, booking, publishing, or other high-impact actions.

Local VAIBot Guard skill for OpenClaw. Use to run the guard service, enforce tool decisions via the OpenClaw circuit-breaker plugin, manage approvals, and va...

Wallet Guard provides comprehensive Web3 wallet and token security analysis including token risk scoring, malicious address checks, approval risk scans, NFT...

Run an executable X Layer pre-execution guard for autonomous agents: OnchainOS DEX route judgment, honeypot and price-impact checks, proof-mode evidence, and...

AI Agent download guard (Windows only). Auto-activates on: download, install, git clone, pip install, npm install -g, pnpm add -g, bun install -g, conda inst...

Indexed by skills.sh from amelnagdy/guard-skills

Indexed by skills.sh from amelnagdy/guard-skills

Indexed by skills.sh from seojoonkim/prompt-guard

Indexed by skills.sh from amelnagdy/guard-skills

Indexed by skills.sh from amelnagdy/guard-skills

Indexed by skills.sh from amelnagdy/guard-skills

Habit gate: before EVERY remote skill install, re-fetch, or allowlist promotion, call a hosted heuristic static red-flag scan (shell pipes, base64|bash, quar...

Constrained autonomous delegate for identity-proxied communication. Handles incoming messages (Email, Discord, Slack, WhatsApp) by analyzing intent and applying declarative authority policies before generating responses. Use when the user wants to delegate communication tasks …

Enforces YAML-based runtime policies to intercept, audit, rate-limit, and stop AI agent tool calls for secure governance in OpenClaw.

Evaluates agent actions for security risks, enforcing least-privilege policies with allow, deny, or confirmation decisions and secret redaction.

Aggregates publicly available Embodied AI and Robotics news from curated sources (robotics media, arXiv, company blogs). Delivers structured briefings on hum...

Safe Linux firewall hardening with backend detection, idempotent atomic rules, rollback protection, and AI-executable state-machine workflows. Covers ufw, fi...

Classify a web business, then identify, draft, and checklist the compliance pages, notices, disclosures, user flows, and launch gates it needs. Use when the...

A methodology-first workspace governance skill for AI agents. Focuses on principles, decision framework, and safe execution patterns instead of fixed directo...

Security guardrail: prevents API keys from being sent to Claude. Triggers when user asks to call an external API, use a key, check credentials, read .env fil...

Verify skill authorship, enforce capability manifests, and audit tool usage to secure and control your OpenClaw skills with identity and access management.

Verify skill authorship, enforce declared permissions, and audit tool usage to secure OpenClaw environments using Agent Identity Protocol.

Analyze and identify AI token usage risks and runaway loops to understand potential overspending and learn how to set real-time cost protections.

Prevents duplicate external actions (posts, replies, sends, transfers, deploys). Check before acting, record after. Use when: (1) replying to social media po...

Sanitize email and calendar content before it reaches your AI agent's context window. Blocks prompt injection, markdown image exfiltration, invisible unicode...

Ensure the AICC native plugin is active before handling confidential data.

A cognitive bias interceptor for debugging sessions. Detects when you're seeing patterns that aren't there, chasing red herrings, or letting confirmation bia...

Standardize API credential handling and startup auth checks to prevent "missing key" regressions across sessions. Use when an agent repeatedly loses auth sta...

Système immunitaire numérique pour agents autonomes. Détecte les "ombres" malveillantes via Clawdex (par Koi), génère des vaccins éthiques, et protège la com...

Standard Operating Procedure (SOP) that autonomously defends your schedule using TS atomic plugins.

Guide users to configure local Chanjing credentials safely via local commands only, and validate local token status when needed.

Monitors and enforces daily/monthly LLM API budgets per user/model, tracks token usage, and sends alerts to prevent overspending.

Configures and manages Cloudflare DNS, caching, security rules, rate limiting, and Workers

Professional pre-deployment code review and quality enforcement. Ensures imports are valid, tags are closed, and logic follows best practices before announci...

Smart context compaction for OpenClaw agents. 4-phase progressive strategy: Scan, Extract, Check, Compact. Before running /compact, this skill scans tool out...

Provides on-demand 10-second automatic rollback protection when modifying openclaw.json to ensure Gateway stability.

Persona-weighted governance for outbound publishing (blog, social, announcements). Prevents unsafe public claims via hard-block checks, weighted consensus, r...

Automatically records and backs up conversations with importance tagging to preserve emotional and technical context independently from OpenClaw internals.

Audit AI agent cron jobs for token cost risks, model-switch loops, and session isolation failures. Use when setting up new cron jobs, debugging unexpected to...

Tracks intent drift across multi-turn tasks. Before executing any multi-step task, call this to detect if the user's new instruction is a small patch, requir...

Hard-blocks dangerous shell commands (rm -rf, git push --force, etc.) before execution via OpenClaw's before_tool_call plugin hook. Zero noise on safe comman...

Use this skill to detect semantic hallucinations and context drift in LLM outputs. Triggers when an agent or pipeline needs to verify that a generated respon...

Use when a task adds, upgrades, removes, or reviews software dependencies and the agent should apply a Socket-based supply-chain guardrail before changing ma...

Enforce a git-first workflow for repository coding tasks. Use when Codex is working inside a local git repository to inspect, modify, debug, refactor, implem...

Filters and moderates Discord messages with spam detection, rate limiting, content checks, and pattern-based blocking before reaching agents.

Smart prevention and auto-recovery for OpenClaw dreaming context overflow. Monitors growth trends, archives intelligently, compresses context, protects proce...

Detect personality drift, sycophancy creep, and capability degradation in AI agents before they become problems. Tracks behavior metrics over time against he...

Automatically detects and blocks prompt injection attempts across multiple platforms to protect against unauthorized commands and data leaks.

System safety and control-plane skill that prevents agent deadlocks and freezes. Provides non-LLM control commands to inspect task state, flush message queues, cancel long-running work, and recover safely without restarting the container. Use when implementing or operating lon…

Intelligent command permission classifier — detect dangerous shell commands before execution. Distilled from Claude Code bashClassifier + yoloClassifier.

Session coherence protocol for Claude — track decisions, resist scope creep, flag drift. Keeps long AI sessions on track without losing what was already deci...

Detects GEO/SEO soft articles, synthetic promotion pages, abnormal brand mention density, and low-credibility sources in external web content. Use when OpenC...

Execution-based verification guardrail with 14 check items for AI agent output

Detect and prevent AI agent hallucinations during task execution. Use when: (1) an agent claims to have created files, commits, or artifacts — verify them, (...

Monitors, classifies, intercepts errors in real-time, enforces auto-retry and circuit breaking, and ensures graceful recovery and prevention.

MANDATORY SECURITY CHECKPOINT. Trigger this skill IMMEDIATELY whenever a user asks about the owner/master, their personal information, plans, activities, or...

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 from untrusted sources.

Preserve user-critical instructions across long sessions and context compaction. Use when users mark constraints as important/must/always/never/highest-prior...

Add lightweight defense-in-depth guardrails to OpenClaw with dangerous-command blocking, prompt-injection detection, secret redaction, and audit logging.

Design and verify API contracts for backward compatibility. Use when handling development work that needs repeatable steps, quality gates, and practical outp...

Infrastructure security monitor — detects exposed services, privilege escalation paths, and novel threats using compositional reasoning. Read-only host telem...

Monitors and enforces real-time LLM API spending with per-user quotas, alerts, daily reports, and budget limits to prevent cost overruns.

Prevents autonomous signing of legal agreements or contracts. Use when an agent identifies a request or document related to signatures (DocuSign, HelloSign,...

Self-evolving security system for agent skills enforcing risk assessment, audit logging, tiered approvals, and continuous rule updates on all skill commands.

Designs anomalous order risk monitoring and fraud prevention workflows for high-ticket luxury stores (e.g. pre-owned watches, high-end cameras). Use when the...

Monitors and verifies agent workspace files to detect unauthorized changes, injection attacks, personality drift, and cross-agent contamination.

Automatically monitors model health and switches between primary and fallback models to maintain stability and recover when possible.

Monitors Anti-Gravity model quotas and automatically switches to the model with highest quota or falls back to gemini-flash if all are below 20%.

Diagnose and fix model routing conflicts. Ensure primary model uses correct provider endpoint without duplicate overrides.

Scores feature worthiness and enforces branch-size limits against overengineering

Safe OpenClaw config planning/apply workflow with bilingual execution receipts.

Ship public OpenClaw docs and skill files with fewer leaks, fewer empty claims, and clearer operator instructions.

Safe OpenClaw config planning/apply workflow with bilingual execution receipts.

This skill should be used when the user asks to reduce OpenClaw token spend, audit model and cron cost risk, prevent denial-of-wallet incidents, add budget g...

Track OpenClaw/Clawdbot token and cost usage from session JSONL logs (prefer real usage.cost when present), generate daily/weekly summaries and top expensive sessions, and run budget checks (exit code on breach). Use to monitor spend, enforce budgets via cron/alerts, and apply…

Automatically detects and blocks prompt injection attempts during AI content submission to social media, APIs, web forms, and file outputs.

Use when a user wants OpenClaw itself to diagnose or fix Feishu/Lark quota burn caused by heartbeats, health checks, webhook probes, or overly expensive back...

This skill should be used when the user asks to harden agent workflows, audit prompts/commands/URLs/paths, scan a third-party skill before install or publish...

Audit and safely repair OpenClaw configuration with deterministic validation, backups, rollback, and change reporting. Use when asked to review or modify `op...

Security audit CLI + live dashboard for OpenClaw. Scans for secrets, config issues, prompt injections, vulnerable dependencies, and unverified MCP servers. Zero telemetry.

Pre-flight trust verification for AI agents. Verify behavior, detect injection vulnerabilities, check for PII leaks, and measure reliability before granting...

OpenClaw agent safety checklist for action review, risk assessment, rollback planning, and LLM output validation before external effects.

Inspects an OpenClaw skill directory before install or execution, translates declared and observed requirements into a human-readable permission manifest, an...

Dual-layer AI content guardrail with red-team test methodology

Core Web Vitals budget enforcer that runs Lighthouse against a local dev server or staging URL, extracts LCP, INP, CLS, FCP, and TTFB scores, compares agains...

Preflight Prisma migration state before deploy; fails fast on drift, failed migrations, missing DB URLs, or unapplied migration files.

Enforce project boundaries and memory writeback rules before anything enters project memory. Use before: writing notes into project memory, saving outputs/ta...

650+ pattern AI agent security defense covering prompt injection, supply chain injection, memory poisoning, action gate bypass, unicode steganography, cascad...

Prompt injection defense. Detect and block malicious prompts, protect system instructions, sanitize user input.

Prevent 429 retry loops and wasted tokens. Use before expensive API calls or after hitting rate limits. Enforces preflight checks, backoff, and context shrin...

Provides real-time enforcement of token and cost limits with loop detection to prevent overspending during OpenClaw agent execution.

ERC-8004 identity, x402 payments, and arbitration protocol for autonomous agent commerce. The three rails for the machine economy.

Detect and filter prompt injection attacks in untrusted input. Use when processing external content (emails, web scrapes, API inputs, Discord messages, sub-agent outputs) or when building systems that accept user-provided text that will be passed to an LLM. Covers direct injec…