Built by builders. Tested in production. New skills added daily.
You need to GetAgentIQ!
Production-ready agent skills built by experts · Trusted by OpenClaw builders worldwide
Individual skills at $4.99 each. Buy exactly what you need.
🇬🇧 UK customers: prices shown ex-VAT. UK VAT at 20% added at checkout where applicable.
Move agent workflows without rebuilding them from scratch. ACP MCP Workflow Adapter turns a source/target agent pair into a vendor-neutral migration recipe with context requirements, MCP/tool mappings, approval gates, unsupported-feature checks, and a practical migration checklist — ideal for teams standardising work across OpenClaw, Codex, Claude Code, Cursor, Gemini, Grok, and other ACP-style runtimes.
Turn any repo into an agent-ready workspace in one safe pass. Repo Project Bootstrap detects the stack, previews AGENTS.md and OpenClaw context scaffolding, preserves existing guidance, and gives developers practical setup/test commands — perfect for clone-to-first-task workflows where agents need clear boundaries before editing.
Find the right OpenClaw skill faster with evidence instead of guesswork. Skill Discovery Curator scans local skill metadata, scores fit against the user's intent, highlights matched terms and review warnings, and returns a ranked shortlist that helps teams choose useful skills before running separate security checks.
Grok Skills Workflow Adapter helps teams turn OpenClaw skills into portable, Grok-style workflow SOPs with compatibility evidence baked in. Instead of manually rewriting instructions for each platform, operators can point it at a skill folder and get a structured workflow, tool list, transfer steps, and audit checks that make cross-agent migration cleaner, faster, and easier to govern.
Rate Limit Handoff Bridge keeps agentic coding work moving when a model, tool, or context window gives out. It packages the essential repo state, blocker, task summary, git evidence, and next actions into a reliable handoff so another agent can pick up the work with less confusion, fewer repeated checks, and stronger safety gates.
Gateway Upgrade Stability Doctor 2026 gives OpenClaw teams a local, read-only upgrade safety gate before gateway changes go near production. It scans operator-provided logs, release notes, and incident evidence for timeout, plugin repair loop, post-upgrade regression, and OAuth-route risks, then returns a clear readiness verdict with redacted snippets and no live gateway command execution.
Audit AI agent permissions, tool access, escalation boundaries, approval requirements, and liability controls before agents touch payments, identity, procurement, benefits, files, or production systems.
Monitor research/intel collection resilience across search providers, rate limits, source freshness, evidence quality, and fallback coverage when Brave, X, Reddit, or other sources are unavailable.
Codex OAuth Route Repair Doctor 2026 is a focused safety check for teams running Codex inside OpenClaw: it turns sanitized route notes and incident snippets into a redacted severity verdict, regression findings, and a practical rollback checklist. It is local-only, read-only, refuses sensitive config files, and is built for the exact moment when an automated repair may have broken OAuth routing and you need evidence before touching production.
Agent Governance Pack helps teams stop agent work moving on vibes. It turns a messy task folder into a clean governance report covering plan evidence, approvals, rollback readiness, tests, and git diff review — with a stable audit ID and a clear proceed-or-hold recommendation for release managers, consultants, and internal automation owners.
Stop guessing which model should handle which task. Multi-Model Routing Skill turns complexity, privacy, latency, provider approval, cost caps, and fallback rules into a clear OpenClaw routing policy — ideal for teams balancing Claude, OpenAI, Gemini, Grok, local models, and ACP harnesses without silent provider drift.
Know whether a skill will actually run before you install or sell it. Skill Compatibility Doctor scans OpenClaw skills for structure, broken Python, missing imports, missing tools, executable-bit expectations, and config assumptions, then returns a clear compatibility verdict with buyer-friendly next steps.
Marketplace Safe Install Scanner gives OpenClaw users a practical quarantine gate for third-party skills: point it at a local skill folder and get a static safety score, pass/review/block verdict, risky-pattern findings, skipped-file evidence, and name-only sensitive-file reporting before install. It never executes target code, calls the network, or reads secrets, making it ideal for cautious buyers, marketplace QA, and teams standardising safe skill intake.
Voice Agent Governance Kit 2026 gives teams a local launch-readiness check for voice agents before pilots become production risk. It scans operator-provided runbooks, policies, and deployment notes for consent, recording, human escalation, safety refusal, accessibility, and audit logging controls, then returns a clear stage recommendation and control-gap list without network calls, subprocesses, or configuration writes.
Long-context AI workflow planning for public-service and regulated teams with token estimates, sensitivity flags, audit controls, and human approval gates.
Design safer Proxmox labs for long-running AI agents with read-only VM/LXC plans, capacity reserves, risk flags, and implementation checklists.
Automated setup of MEMORY.md, daily memory files, HEARTBEAT.md. Bootstrap your agent in minutes.
Monitor session bloat, detect context overflow, alert before crashes. Keeps your agent healthy 24/7.
Maintain context across sessions with smart chunking and lazy-loading. Never lose your agent's memory again.
Intelligent browser automation with anti-bot bypass, snapshot caching, and state prediction.
Real-time funding rate scanner. Finds negative rates for profitable long positions. Binance futures support.
Track TRON/ETH/BSC holdings, analyse positions, monitor yields and gas fees.
Professional trading strategy guides for prediction markets and crypto. Risk management and trend analysis.
Budget tracker enforcing cost caps with auto-failover chain. Tracks spend, selects safest model given budget.
Microsoft Dynamics 365 Business Central connector. Query financials, post journals, sync data.
Command-line trading interface for Kalshi prediction markets. Place, monitor and manage positions.
Kraken exchange connector with safety guards. Query balances, place orders, monitor positions.
Unified OAuth for 1000+ apps. Authenticated access to Gmail, GitHub, LinkedIn, Slack without managing API keys.
Route agent outputs to the right messaging platform. Handles Telegram, Discord, Email, Slack, Teams.
Meta-skill for dynamic skill composition. Chain and orchestrate other skills as sub-tasks.
Execute tasks in isolated cloud environments. Parallel execution with result aggregation.
Analyses conversation quality, identifies improvement opportunities, continuously optimises response strategies.
Complete first-run setup for OpenClaw. Guided wizard takes new users from fresh install to fully configured agent in 5 minutes.
Morning digest of overnight agent activity, cron results, memory highlights, and today's agenda.
Continuous skill health checks, performance baselines, automated alerts on degradation.
AES-256 encrypted local vault for API keys, tokens, secure exports, and access audit logs.
Visual cron timeline, conflict detection, schedule validation, and per-job error history.
Benchmark first-run setup friction, audit installs, and score onboarding completeness.
Comprehensive security audit tool for OpenClaw skills. Scans for malicious patterns, data exfiltration, and supply-chain risks with 37 detection patterns.
Track, analyze, and optimize OpenClaw API spending across all providers. Cut costs 50-90% with smart model recommendations.
Establish performance baselines, track metrics over time, detect regressions automatically.
Reinforcement learning feedback loop. Train lightweight adapters on successful task sequences.
Real-time memory health monitoring with automatic backup and repair. Tracks context window fill rate, detects compaction-related data loss, creates automatic backups.
End-to-end Microsoft Teams enterprise deployment automation. Generates bot manifest, channel routing rules, security policies, and welcome cards in minutes.
Bi-directional bridge between OpenClaw and n8n workflow automation. Trigger complex workflows, receive webhooks. Includes 8 pre-built templates.
Store, search, version, and reuse prompt templates across sessions and agents. Enterprise-grade prompt management for consistent, reusable AI interactions.
5 reusable agent archetypes — Researcher, Writer, Executor, Analyst, Coordinator — with pre-configured skills, memory templates, and SOUL.md for rapid swarm deployment.
Chain OpenClaw skills into automated, multi-step workflows. Define pipelines where skill A feeds skill B, with conditional branching, error handling, and output routing. Perfect for repeatable tasks like 'run audit → check memory health → summarize costs' in one orchestrated command. Pre-built templates for daily audit, full scan, memory maintenance, cost review, and security hardening workflows.
Expose OpenClaw tools as MCP endpoints. Connect to Claude Code, n8n, and any MCP-compatible platform. Auth, rate limiting, n8n templates, and audit logging built in. Security-first: exec permanently blocked from exposure.
Benchmark 18 LLM models across 6 providers on cost, quality, and latency. Task-type recommendations (coding/analysis/creative), three-tier output (best/budget/premium), and CSV/JSON export. Stop guessing which model to use — measure it.
Auto-compress daily memory files into weekly summaries. Detect duplicates, archive stale content. Reduce memory bloat by 60-80% without losing context. Dry-run preview before any changes.
Map your entire skill ecosystem. Scan 44+ skills, detect orphans, score ecosystem health (A–F), get complementary skill suggestions, and export GraphViz/JSON graphs. Know exactly what you have and what to add next.
OpenClaw-specific security scanner: audits gateway config, detects CVEs, scans skills for prompt injection, generates A-F security scores with auto-remediation.
Automated Hermes Agent → OpenClaw migration. Imports configs, maps 17 Hermes tools to ClawHub equivalents, generates feature comparison benchmarks.
Complete first-30-minutes setup wizard for OpenClaw. From fresh install to fully configured agent — no JSON editing required.
Real-time multi-agent cost tracking. Monitor API spending, per-model efficiency breakdowns, budget alerts, and recommendations to cut costs 30-50%.
VirusTotal for OpenClaw skills — static analysis trust scanner. Audits skill code for exfiltration, shell injection, eval/exec abuse. A-F trust grades.
Intelligent memory management that keeps your agent's knowledge intact while reducing bloat. Performs zero-loss compaction (merges duplicates while preserving critical context), semantic search across all memory files, and auto-tagging for instant retrieval.
Transform experience into expertise. Agent Learning Loop records task outcomes, identifies patterns, and suggests optimal approaches based on history. Your agent gets smarter with every execution, building an experience database that compounds over time.
Turn OpenClaw into a microservice platform. Expose skills as MCP endpoints that Claude Code, n8n, and Make can call. Auth, rate limiting, and audit logging built in.
Best-of-both-worlds agent stack: OpenClaw orchestration + Hermes self-repairing coding engine. Route heavy coding to local Gemma 4 — zero API cost, autonomous error correction.
Always-on security layer detecting prompt injection, jailbreaks, tokenade attacks, and data exfiltration with 19 threat patterns across 5 attack categories. Auto-quarantine and instant alerts.
DeFi portfolio manager for OpenClaw. Scan live yield opportunities across Solana, Ethereum, Base, Sui via DeFiLlama. Risk-adjusted APY rankings across 21 protocols with rebalancing recommendations.
One-command security scan and auto-fix for OpenClaw. Scans config, memory, git history for 40+ API key patterns; checks gateway auth, TLS, file permissions, port exposure. 0-100 security score with auto-fix mode. Covers CVE-2026-33579.
AI-powered trading with safety-first architecture. Connect to Binance, Hyperliquid, Bluefin, Polymarket, Kalshi. Five built-in strategies (DCA, grid, momentum, funding rate arb, mean reversion). Defaults to paper mode — no autonomous trading without explicit --live flag.
Git-style version control for agent memory. Snapshot before compaction, diff to see what changed, rollback to any point. Loss-detection algorithm identifies removed dates, names, decisions and reports preservation ratio per file. Auto-backup on 6h cron schedule.
Auto-detects your hardware (GPU VRAM, RAM, CPU), recommends the best Ollama-compatible models, and configures intelligent hybrid cloud/local routing. Supports Gemma 4, Llama 3.1, Mistral, Qwen, Phi. Cut API spend 50-90%.
Watchdog that keeps your OpenClaw agent running without manual intervention. Monitors cron jobs, memory files, config, sessions, skills, and network health — detects failures, diagnoses root causes, applies the safest fix automatically.
Automatically classifies every task by complexity and routes it to the cheapest model that can handle it, saving 60-80% on API costs. Four complexity tiers, live dashboard, daily and monthly budget caps.
Tamper-evident audit trail, RBAC, SSO templates for Okta, Azure AD, Google Workspace, Auth0, and OneLogin. Compliance report generates scored assessment for enterprise IT sign-off. The #1 blocker for enterprise adoption, solved.
Five-level fallback chain — rotating user agents, session cookies, headless Chromium, 24h cache — until it gets the content or tells you exactly why it couldn't. Tracks success rates per domain and learns the optimal level over time.
Extract every factual claim — numbers, names, causals, dates — and cross-reference against your source documents. Verified claims get citations; contradicted claims get flagged. 0-100 trust score with automatic routing for human review.
Orchestrator handles smart decisions while cheap workers handle volume. Eight task categories, auto-escalation when a worker fails, full cost logging with Opus-baseline comparison. Save 70%+ on API costs from day one.
When your primary LLM provider goes down, your pipelines keep running via automatic failover. Real-time traffic-light health monitoring, budget guards during failover, and one-command state snapshot/restore. No more lost work from mid-run outages.
The complete first-run setup experience OpenClaw ships without. 6 validated steps — API keys, model selection with budget profiles, memory setup, first skill install, and channel connection. Progress saved so interruptions don't cost you ground. Fastest path from fresh install to working agent.
Turn your inbox from a source of anxiety into a prioritised action list in seconds. Connect Gmail or Outlook via IMAP, triage 24 hours of email into four tiers with confidence scores, track unanswered follow-ups, and draft replies in your style — presented for approval, never auto-sent.
Describe any web task in plain English and xAI's Grok vision-action loop handles every click, scroll, and extraction automatically. No selectors, no bespoke scrapers, no brittle XPath. Built-in simulation mode lets you test full pipelines before spending a single token. Schedule recurring research tasks via OpenClaw cron and chain JSON output into memory or reporting skills.
Transform exported OpenClaw memory into a clean import-ready JSON payload, preserve useful metadata, and redact selected fields like email, phone, or tokens before handoff. Ideal for privacy-first teams building their own memory layer.
Checks timeout budgets, retry coverage, partial-output handling, and operational evidence in one pass, then returns a clear risk score with prioritised fixes. Built for ClawHub operators scaling retrieval workflows.
Strips filler words, counts repeated demand terms, and gives you a lightweight signal on where the strongest pull is. Perfect for marketplace creators, solo founders, and product teams staying close to real user demand.
A fast, lightweight way to get a quick read on a website without building a full scraping stack. Perfect for marketers, researchers, and builders who want useful web context in seconds.
Removes friction from cross-agent delegation by packaging prompt and context into one structured, repeatable handoff that is easy to automate and easy to trust.
Reviews each workflow step for timeout budgets, retries, fallback paths, and hung-run detection, then scores the flow with practical hardening recommendations for API, scraping, and retrieval pipelines.
Shows exactly where a long-running process should emit something valuable sooner, helping teams improve resilience and user experience where dark waits are unacceptable.
Returns a clear verdict with concrete reasons and next actions, turning vague concern into measurable signals for operators running serious production agents.
Gives teams a straightforward observability score with fix recommendations, ideal for faster debugging, shorter incident response, and stronger confidence in retrieval-heavy automations.
Ideal for consultants, internal reviewers, and operators wanting a lightweight security baseline before shipping. Returns ranked findings with concrete remediation guidance.
A simple but powerful optimisation tool for builders serious about model-facing interfaces. Keeps your core contract intact while eliminating unnecessary bloat from every call.
Perfect for teams running production Telegram, Discord, or Slack automations. Recommends the next stabilisation moves with evidence before users start noticing silent degradation.
Analyze sample voice interactions for intent, urgency, next actions, and escalation risk, then generate practical starter assets for review before connecting live channels.
Generate complete voice agent workflow specs from parameters — no transcripts needed. Specify use case, industry, languages, and tools; get call stages, system prompt scaffold, tool contracts, escalation triggers, QA scenarios, and launch checklist instantly.
Validate whether a crawler run is actually useful before teams trust the output. Checks same-domain coverage, duplicate suppression, relevance scoring, failure rates, and thin-content drift, then returns severity-ranked fixes.
Find token bloat in MCP and tool schemas before it slows every agent call. Ranks the heaviest schema nodes, flags budget risk, and recommends safe trims that preserve the core integration contract.
Create privacy-safe debug bundles by redacting secrets, tokens, credentials, emails, hostnames, IPs, and private paths before sharing diagnostics.
Mine GitHub issues, discussions, and exports for repeated pain points, feature demand, bug clusters, content ideas, and build-brief priorities.
Read-only OpenClaw security scanner that finds exposed secrets, insecure permissions on sensitive files, and risky skill code patterns, then returns a 0–100 evidence-backed security score.
Run safe OpenClaw upgrade and rollback readiness checks with backup evidence, version posture, smoke tests, and rollback guidance.
Score every skill for quality, surface stale or risky skills, detect duplicate clusters, and generate an approval-gated curation plan. Advisory-first — no silent deletions, no unapproved changes.
Diagnose OpenClaw and Hermes setup issues across binaries, paths, stale tool schemas, gateway reloads, services, workspace health, and channel readiness.
Turn scattered governance notes into a clear enterprise-readiness review for AI-agent deployments — checking identity, config, sandboxing, permissions, audit, retention, incident response, secrets, and human oversight.
Make any repo agent-ready without guessing — read-only stack detection, setup command recommendations, AGENTS.md check, and sensitive file flagging for a safe handoff before OpenClaw agents start work.
Make any repo agent-ready in minutes — read-only stack detection, setup guidance, AGENTS.md gap checks, and sensitive-file flagging for safe agent handoff before touching a new codebase.
Know before you ship. Checks whether a repo has the six safety controls agents need before running autonomously in CI — returns a clear READY or BLOCKED verdict with per-control coverage detail.
Pre-install risk gate for OpenClaw marketplace skills — scored INSTALL_OK / REVIEW_BEFORE_INSTALL / DO_NOT_INSTALL verdict with severity-ranked findings and secrets automatically redacted.
Validate your routing policy before real traffic hits it. Checks privacy floors, cost caps, tool-calling requirements, and approved-model restrictions — per-task verdicts with primary model, fallback chain, and BLOCKED reasons.
Governance readiness for councils, NHS, FOI, and regulated services deploying AI agents. Eight controls assessed, GO_WITH_CONTROLS or BLOCKED verdict, starter SOP template. Use --regulated for stricter thresholds.
Pre-task environment scan that checks workspace state, tool availability, config integrity, and memory headroom — returns a clear GO / HOLD / ABORT verdict before your agent commits to a high-stakes run.
Turn Windows setup friction into a clear diagnostic and action plan. Covers WSL friction, prerequisites, shortcut manifests, admin vs. normal-user steps, and remediation guidance without touching config.
Multi-Agent Routing Control Plane is the control tower for serious OpenClaw teams running more than one worker. It audits route tables and handoff folders for missing owners, invalid statuses, absent next steps, weak evidence, and unclear…
Integration Contract Auditor Pro is the pre-publish preflight check every OpenClaw skill and integration repo needs before going to Rooster or ClawHub. One command scans for missing contract files, broken SKILL.md script references,…
Voice Agent Governance Kit gives you a structured, evidence-based governance readiness report for any voice-agent or contact-centre deployment in one command. Scan your policy documents, call-flow configs, and compliance notes for the…
Gateway Upgrade Stability Doctor is the read-only smoke checker that tells you whether your OpenClaw gateway and workspace survived an upgrade before you resume any cron jobs, agent pipelines, or production automations. One command checks…
Marketplace Skill Risk Scanner Pro is the pre-install vetting tool for serious OpenClaw operators. Deep static analysis flags remote script execution, destructive commands, embedded secrets, dynamic code execution, privilege escalation,…
Local Model Task Router stops the expensive habit of defaulting to one model for everything. Describe any task, tell it your hardware tier and priority (speed/balanced/quality/privacy), and get a specific local model recommendation from a…
Voice Contact Centre Agent Kit generates a complete, compliance-ready voice-agent pilot pack in one command. Supply your sector, channel, and intent list and get back a structured JSON artifact covering call flows, sensitive-intent…
Analyze OpenClaw session exports for cost drift, retry loops, model mix, expensive turns, and optimization opportunities without exposing private logs.
Stop overspending on routine AI work. Tell it the task, complexity, privacy level, context size, and remaining daily budget, and it returns a practical model route: local-first, balanced, premium, or local-only when budget is gone.
Batch routing for entire task backlogs. Define tasks and model catalogue as JSON files, get routing decisions and total paid spend estimate before any model is called. Config-as-code approach for version-controlled routing policies.
Audit integration contracts, API specs, webhook payloads, environment expectations, and handoff docs for blockers before automations fail.
Pre-install trust scanner for OpenClaw skills: hidden network calls, credential harvesting, shell execution, unsafe file access, and supply-chain risk.
Review long-running autonomous loop plans for cadence, retry settings, stop conditions, escalation rules, and human review checkpoints before they go live.
Generate practical voice-agent workflow templates for support, sales, intake, bookings, casework, and back-office operations.
Monitor public research sources, docs, competitor pages, and pricing pages for compliant changes with snapshots, hashes, bounded diffs, and evidence trails.
Pre-flight trust scoring and governance for browser and computer-use agents. Define your governance policy once and audit any planned workflow in seconds with a 0–100 trust score and ALLOW / REVIEW / BLOCK verdicts.
Generate structured multi-step agent workflow templates for admin, casework, research, support, and custom processes — YAML contracts, operator playbooks, and test plans with human approval gates and evidence fields.
Build structured browser-agent workflow blueprints with human approval gates, evidence requirements, error handling, and handoff schemas before any live browser automation runs.
Convert raw call-centre transcripts and voice-agent notes into structured operational triage with case summary, sentiment, urgency scoring, recommended queue, and follow-up actions.
Generate ready-to-review dashboard extension scaffolds with an extension manifest, frontend tab stub, backend route stub, and validation report for custom tabs, plugins, and agent operations panels.
Recommends the right local quantization strategy and fallback plan before spending on cloud APIs or moving sensitive work off-device — based on VRAM, RAM, CPU, task complexity, and privacy level.
Hermes Migration Retention Doctor helps OpenClaw teams spot why a user or workflow is at risk of switching away, then turns messy complaints into a clear rescue plan. It classifies churn drivers like reliability, missing capabilities, UX friction, integrations, trust, speed, and cost, then recommends practical fixes your product or ops team can act on before the account is gone.
Browser Crawler Research Verifier turns raw crawl exports into defensible research evidence. It checks whether collected pages actually support your claim, flags weak citations, stale sources, duplicate URLs, invalid links, and tracking-heavy URLs, then produces a clean source hygiene report for analysts, content teams, and agent workflows that need proof rather than scrape volume.
AI Signal Provenance Auditor is a pre-publish trust gate for AI-generated market, crypto, trading, and intelligence signals. It checks every signal for source lineage, model/run context, human review, disclaimers, freshness, confidence basis, advice-style wording, and brand authorization so teams can stop unsupported calls before they become a compliance or reputation problem.
Intelligent token budget management with automatic model downgrade for OpenClaw agents. Tracks per-agent and per-flow API spend in real-time, auto-downgrades to cheaper models for simple queries (FAQ, greetings, status checks), enforces daily/monthly budgets with hard stops, and provides cost attribution by flow/agent. Uses OpenClaw's before_agent_reply hook for zero-latency query classification. Use when asked to "a
Monitor and audit agent reliability using heartbeat logs, task runs, retry policies, stale-session signals, missed checkpoints, and escalation rules. Use when agents appear narcoleptic, fail to finish tasks, silently stall, or need trust infrastructure around automated work.
Browser-side agent session protection against the ClawJacked vulnerability and prompt injection attacks. Monitors agent session files for unauthorized access, detects prompt injection patterns in incoming messages, validates session origins, and provides session isolation hardening. Use when: (1) protecting against ClawJacked browser-based agent takeover, (2) detecting prompt injection in incoming messages before pro
Playwright-based scraper with rotating user agents, residential proxy support, human-like delays, and CAPTCHA detection. Handles sites that block standard scrapers.
Production-ready OpenClaw skill package for API Credential Guardian.
Validate Codex/OpenAI provider routes after OpenClaw doctor repairs, especially OAuth-only GPT-5.5/Codex setups where route rewrites can break working auth. Use before running doctor --fix or after route/auth regressions.
Production-ready OpenClaw skill package for Composio Wrapper Skill.
Secure API key and credential management. Validates keys on startup, rotates on failure, alerts when credentials expire or are revoked.
Monitors all OpenClaw cron jobs for persistent errors, silent failures, and consecutive error streaks. Auto-diagnoses root causes, applies known fixes, and alerts when crons go dark. Prevents the "11 failing crons" pattern.
Production-ready OpenClaw skill package for Skill: cron-maintenance-window-guard.
Production-ready OpenClaw skill package for Cron Timeout Circuit Breaker.
Automated CVE detection, patching, and supply-chain trust verification for OpenClaw agents. Monitors NVD and GitHub advisories for new CVEs affecting OpenClaw or installed skills, generates and applies patches, validates skill trust chains, and alerts on supply-chain attacks. Use when: (1) checking for new CVEs affecting OpenClaw installation, (2) auto-patching known vulnerabilities, (3) verifying skill supply-chain
Production-ready OpenClaw skill package for daily-skill-auto-installer.
Nightly memory consolidation and smart recall for OpenClaw agents. Mimics human sleep-based memory processing: clusters related memories, summarizes old daily files, extracts key learnings, prunes redundancy, and builds a searchable knowledge graph. Use when: (1) consolidating scattered daily memory files into coherent long-term knowledge, (2) finding forgotten context across old memory files, (3) reducing memory fil
Vendor-neutral model and workflow routing for enterprise agent platforms across OpenClaw, Grok, Gemini Enterprise Agent Platform, Hermes, local models, browser agents, and voice workflows. Use when choosing model/vendor execution lanes based on task type, data sensitivity, latency, cost, tool needs, compliance, and fallback policy.
Auto-classifies "unknown" failures in the Sparta swarm by scanning log output, error messages, and agent memory files for known failure signatures. Turns vague "unknown" errors into actionable categories (scrape_block, rate_limit, timeout, auth_error, empty_result, model_error) with suggested fixes. Reduces the dominant 24-34x/week "unknown" failure bucket in failure-analyst reports.
One-click setup for running Google Gemma 4 models locally with OpenClaw via Ollama. Auto-detects hardware (VRAM, RAM, CPU), recommends the optimal Gemma 4 variant (E2B/E4B/12B/27B), installs and configures Ollama, pulls the model, sets it as an OpenClaw provider, and verifies the connection. Use when asked to "set up Gemma 4", "run local model", "install Gemma 4", "configure Ollama for OpenClaw", "set up local AI", "
Production-ready OpenClaw skill package for getagentiq-auto-updater.
Production-ready OpenClaw skill package for getagentiq-weekly-updater.
Automated incident response toolkit for compromised OpenClaw agents. Provides forensic snapshot, session quarantine, credential rotation checklist, timeline reconstruction, and recovery procedures. Use when: (1) an agent has been compromised or shows signs of takeover, (2) responding to a detected security breach, (3) performing post-incident forensics, (4) rotating credentials after a compromise, (5) generating inci
Detects LLM authentication failures, rejected requests, and model unavailability then automatically fails over to an alternative model. Prevents cascading failures when the primary LLM endpoint is misconfigured or unavailable.
Production-ready OpenClaw skill package for LLM Resilience Guard.
Build cost, quality, privacy, and failover policies for multi-provider LLM routing across OpenAI, OpenRouter, local models, Hugging Face, custom endpoints, and other providers. Use when avoiding model lock-in or controlling spend.
Orchestrate multiple OpenClaw agents in parallel or sequential pipelines with dependency management, result aggregation, and failure handling. Use when: (1) running 2-16 parallel agent tasks and aggregating results, (2) building multi-step agent pipelines where each stage depends on prior output, (3) distributing large tasks across specialist sub-agents (research, writing, coding, analysis), (4) managing fan-out/fan-
Validate Codex/OpenAI OAuth routing before and after doctor/repair commands, especially GPT-5.5 route drift, wrong provider aliases, and broken OAuth-vs-API-key configuration.
Production-ready OpenClaw skill package for OpenClaw Quick-Start Skill Pack.
Resilient multi-layer web scraper with automatic failover. Use when scraping web content that may be JS-rendered, behind bot protection, or on sites that block standard HTTP requests. Supports three-tier failover chain: Brave Search API, then Bright Data Web Unlocker (with optional browser render mode), then Playwright headless browser. Use for job boards (LinkedIn, Glassdoor, Reed, Indeed, CWJobs, TotalJobs), social
Agent health monitoring, failure detection, and auto-restart for OpenClaw agents running 24/7 in production. Monitors gateway status, skill loading health, API key validity, session memory, response latency, and cron job success rates. Sends alerts on failures and optionally auto-restarts degraded components. Use when: (1) running OpenClaw agents in production and need uptime monitoring, (2) diagnosing gateway failur
Use when asked to create client-ready Word documents, reports, interview templates, capture templates, assessment matrices, user guides, client memos, or any .docx output where markdown conversion artefacts would look amateur. Builds polished DOCX files directly with docx-js.
Production-ready OpenClaw skill package for Prompt Injection Hardening Skill.
Generate safe public-service agent workflow templates for FOI triage, casework, form intake, briefing notes, document generation, RAG/search review, and admin queues with guardrails for privacy, human review, auditability, and citizen-facing quality.
Intelligent rate limit handler with exponential backoff, queue management, and per-provider limits. Prevents 429 errors across all API calls.
Production-ready OpenClaw skill package for Reddit Poster — Post to Reddit Subreddits.
Production-ready OpenClaw skill package for Resource Validator & Fallback Manager.
Generic retry decorator with exponential backoff, jitter, and configurable retry conditions. Works with any function.
Production-ready OpenClaw skill package for Scheduled Tasks Showcase Skill.
Comprehensive runtime security for OpenClaw agents — skill trust chain verification, CVE auto-alerting, runtime behavior monitoring, and kill-switch capability. The 'antivirus for OpenClaw.' Use when: (1) verifying installed skills haven't been tampered with, (2) monitoring agent runtime for suspicious behavior (data exfil, unexpected network calls, privilege escalation), (3) checking for new CVEs affecting OpenClaw
Production-ready OpenClaw skill package for Self-Tuning RL Hook.
Monitors OpenClaw session file sizes and automatically triggers pre-restart state dumps before sessions exceed safe thresholds. Prevents context loss from bloated sessions that cause timeout cascades.
Production-ready OpenClaw skill package for session-recovery-cost-audit.
ClawHub marketplace skill verification and malware detection. Scan any skill before or after installation for malicious code, supply-chain risks, suspicious behaviors, and known-bad signatures. Generates trust scores and detailed security reports. Use when: (1) evaluating a ClawHub skill before installing, (2) auditing all installed skills for hidden threats, (3) checking if a skill has been compromised, (4) verifyin
Production-ready OpenClaw skill package for skill-security-validator.
Production-ready OpenClaw skill package for Sunday AM Cron Guard.
Recover abandoned or half-finished agent work by capturing git diffs, changed files, terminal/log excerpts, TODO markers, and creating a clean continuation note. Use when tasks stall, agents stop mid-flow, work is unfinished, or a handoff/resume brief is needed.
A bundle of 20 production-ready micro-skills for AI-driven development, trading, finance, and productivity automation. Covers crypto signals, Monte Carlo simulations, Git hygiene, HTML parsing, prompt A/B testing, job scraping, LinkedIn content generation, and more. Built for OpenClaw agents.
Test REST APIs by sending requests, handling authentication, parsing responses, and flagging errors. Use when validating API integrations, debugging endpoints, or checking that credentials and rate limits are working. Supports Bearer token, API key, and Basic auth.
Monitor OpenClaw cron job health, flag genuine errors, and suppress known false positives like quiet-hours skips and message-tool misfires. Use for daily swarm health checks. Returns errored jobs, success rate, and improvement proposals.
Monitor crypto funding rates and z-score signals across pairs to identify mean-reversion and momentum opportunities. Use when scanning for paper or live trading signals on BTC, SOL, ETH, and alt coins. Returns STRONG/MODERATE/WEAK signals with annualised yield estimates.
Compile a structured daily briefing from emails, RSS feeds, memory signals, and portfolio data. Use when starting the day with a consolidated view of overnight signals, market moves, job opportunities, and AI/tech news. Produces a scannable Telegram or markdown digest.
Calculate Dollar Cost Averaging (DCA) entry points, cost basis, break-even prices, and position sizing for crypto or stock investments. Use when planning a DCA strategy, reviewing existing DCA positions, or calculating how many buys are needed to average down to a target price.
Analyse emails for investment signals, job opportunities, AI business intel, and macro/geopolitical alerts. Categorises emails by type and extracts actionable data points. Use when processing newsletters, job alerts, or any email corpus for signal extraction.
Validate a .env file by checking for required keys, flagging missing or placeholder values, and detecting potentially expired credentials by pattern. Use before deploying an agent or running a pipeline to ensure all required environment variables are present and non-empty.
Auto-commit uncommitted workspace files, run git gc for repo compaction, and report repository health. Use as part of nightly maintenance to ensure no research or code is lost, repo stays lean, and git history is clean.
Extract structured data from HTML using CSS selectors and regex patterns. Use when scraping web pages for specific fields like titles, prices, links, tables, or any structured content. Returns clean JSON or plain text.
Scrape job listings from Reed, Indeed, LinkedIn, and Glassdoor using Bright Data's job_search_scraper zone. Returns structured job dicts with title, company, location, salary, and URL. Use when searching for jobs, monitoring hiring trends, or building job alert pipelines.
Generate engaging LinkedIn posts from a topic, URL, or brief. Use when creating thought leadership content, sharing project updates, or building personal brand on LinkedIn. Produces hook-led posts with structured body, CTA, and relevant hashtags optimised for LinkedIn algorithm.
Auto-archive old MEMORY.md entries when the file exceeds a size threshold. Use when MEMORY.md grows too large and risks bloating the agent context window. Archives entries older than N days to monthly archive files, preserving a rolling summary.
Run Monte Carlo portfolio simulations to estimate future value distributions, success probabilities, and risk scenarios. Use when stress-testing a portfolio against retirement targets, big purchase goals, or long-term financial planning. Configurable for any mix of crypto and stocks.
Fetch live prices for stocks and crypto assets via Financial Modeling Prep (FMP) API with Yahoo Finance fallback. Use when you need current market prices for a portfolio, Monte Carlo simulation, or daily tracker. Returns a dict of symbol to price.
A/B test two prompt variants by running both against a judge model and scoring on clarity, completeness, efficiency, and safety. Use when optimising agent prompts, cron job instructions, or system prompts. Returns scores, winner, and improvement notes.
Fetch RSS feeds and write signal summaries to MEMORY.md. Use when monitoring news sources, blogs, or any RSS/Atom feed for relevant signals. Filters by keywords and appends structured summaries to the agent's MEMORY.md file.
Validate JSON or YAML data against an expected schema with clear, actionable error messages. Use when debugging API responses, validating config files, or ensuring agent output matches expected structure before passing to downstream tools.
Three-tier resilient web scraper with automatic failover: Brave Search API → Bright Data Web Unlocker → Playwright headless browser. Use when scraping any URL that may block direct requests, needs JS rendering, or requires maximum reliability. Never returns empty — always escalates through tiers until content is retrieved.
Fetch and summarise YouTube video transcripts via TranscriptAPI or yt-dlp fallback. Use when researching YouTube content, building a knowledge corpus from channels, or extracting verbatim quotes from videos. Returns full transcript text or structured summary.
Filters and validates YouTube transcript data to prevent transcript content fragments from being misclassified as swarm failures. Pre-processes YouTube insights before ingestion into the failure analyst pipeline.
Agent Financial Action Risk Gate gives finance and operations teams a practical safety check before an AI agent can touch payments, vendors, payroll, identity, approvals, or other high-impact workflows. Feed it a sanitized action manifest and it returns a plain-English PASS/WARN/BLOCK result with missing controls and deployment recommendations — ideal for turning risky automation ideas into auditable release decisions.
Codex Route Config Diff Guard gives OpenClaw operators a focused pre-flight check for the config changes that matter most: Codex routes, OAuth markers, model/provider mappings, and rollback risk. Compare sanitized before/after snapshots and get a redacted PASS/WARN/BLOCK decision before a subtle route rewrite turns into an outage.
Gateway Regression Smoke Runner helps teams move from “the gateway feels broken” to evidence-backed go/no-go decisions. Drop in sanitized logs, status snippets, or upgrade notes and it flags outage, latency, plugin repair, OAuth route, and rollback signals with recommended next checks — perfect for safer OpenClaw gateway upgrades.
Intel Quota Fallback Planner keeps research and monitoring pipelines honest when primary sources hit quotas, blocks, or stale data. It ranks available fallbacks, recommends retry/backoff behaviour, and labels evidence confidence so teams can continue coverage without pretending weak source coverage is strong.
Voice Agent Escalation Logger gives voice teams a clean way to convert risky calls into redacted, review-ready escalation evidence. It spots urgent, sensitive, and friction triggers, masks common contact details, and creates a concise handoff summary for human queues — ideal for safer voice assistant governance and contact-centre operations.
Agent Permission Audit Kit is a fast governance checkpoint for autonomous agents with risky authority. It spots payment, procurement, identity, production-write, and personal-data permissions, then tells operators whether the agent is low risk, approved with controls, or blocked until human approval, scoped limits, and audit logging are in place.
Codex OAuth Route Repair Doctor gives OpenClaw admins a safe way to diagnose Codex/OAuth route breakage before anyone edits config. It redacts credentials, identifies risky API-key route substitutions, and produces a rollback-aware repair checklist so teams can fix provider routing deliberately instead of guessing under pressure.
Gateway Upgrade Stability Doctor is a local-only preflight for risky OpenClaw gateway upgrades. It classifies logs and notes into proceed, review, or block decisions, highlights upgrade-breaking signals, redacts secrets, and gives operators the rollback controls they need before changing production infrastructure.
Search Intel Fallback Monitor keeps research workflows honest when sources degrade. It turns outage, quota, and partial-result notes into a health rating, fallback plan, and evidence policy so agents do not overclaim weak findings or publish fallback data as primary confirmation.
Voice Agent Governance Kit gives teams a practical launch-readiness audit for customer-facing voice agents. It checks the controls that matter — consent, escalation, transcript retention, monitoring, accessibility, and scope boundaries — and produces a clear production, pilot, or block verdict before a voice workflow goes live.
Security review for OpenClaw skills in one command: Skill Security Audit Pack scans skill folders for credential harvesting, hidden POSTs, unsafe execution, obfuscation, risky install scripts, binaries, and missing metadata, then produces a clean PASS/FAIL gate with JSON evidence your release team can keep. Built for marketplace operators, QA agents, and enterprise teams that need buyer-ready security confidence before a skill goes live.
Before an Office agent touches spreadsheets, emails, PDFs, decks, or RAG outputs, run Agentic Office Governance Kit. It converts a manifest into a buyer-friendly governance report with approval, audit logging, retention, redaction, DPIA, data residency, and rollback checks — ideal for UK/EU teams that want automation speed without losing control.
Enterprise buyers ask the same hard questions: SSO, RBAC, audit logs, DPA, subprocessors, data residency, retention, incident response, support, security evidence, and backup/restore. Enterprise Agent Procurement Readiness gives agent teams a fast, structured gap report so procurement blockers are visible before they slow a deal down.
AI compute is no longer just a technical question — it is a trust, permitting, energy, water, emissions, community, and reputation question. Agent Compute Legitimacy Monitor turns infrastructure evidence and public claims into a clear risk report so operators, investors, and advisors can identify legitimacy gaps before they become expensive headlines.
ACP Router gives OpenClaw teams a simple, auditable way to decide where work should run: native tools, browser automation, coding sessions, one-shot ACP runs, or persistent harness sessions. It recognises explicit Claude Code, Cursor, Gemini, Codex ACP and other harness requests, flags risky production or credential work for approval, and returns clean JSON routing guidance operators can trust.
Agent Channel Permission Auditor helps teams stop permission sprawl before it becomes a production incident. Point it at channel notes, configs, scripts, or exported logs and it returns a concise JSON audit of risky access, missing ownership evidence, and next-step remediation checks across messaging, repo, filesystem, billing, and deployment surfaces.
Gateway Stability Repair Doctor is a first-response triage kit for OpenClaw gateway trouble. It scans relevant notes, logs, and configs, flags repair-risk patterns, and produces a structured checklist covering gateway health, dependency churn, startup symptoms, and missing regression evidence.
Low-Power Self-Hosting Doctor helps OpenClaw users run lean on Raspberry Pi, ARM, mini PC, and small VPS deployments. It reviews setup notes and configs for resource mismatches, swap/disk pressure clues, heavy dependencies, fragile services, and missing lightweight fallback plans.
OpenClaw Security Hardening Pack is a practical pre-release security checklist for agent workflows and skills. It scans documentation, scripts, configs, and logs for secret exposure clues, unsafe shell usage, overbroad privilege hints, writable-config risk, and missing rollback evidence, then returns a structured remediation report.
Persistent Agent Governance Kit gives long-lived agents a governance safety net. It scans notes, configs, and workflow evidence for retention, consent, revocation, identity, channel ownership, and audit-trail gaps so teams can deploy persistent memory and cross-channel agents with clearer controls.
Every bundle is hand-picked — the right skills, in the right combination. No guesswork.
3 skills bundled together: Outage Resilience Pro · Onboarding Wizard Pro · Email Triage Agent
Save vs buying individually — all three skills in one purchase.
Supercharge your subscription with these exclusive tools — only useful if you're a subscriber.
Find your skill, pay once ($4.99) or subscribe ($17.99/mo). Bundles available for bigger savings.
Instant delivery. Individual .py and SKILL.md files (no zip). Subscribers enter token once — stored in your browser.
Drop files into your OpenClaw skills/ folder. Follow the release notes. Running in minutes.
Tutorials, strategy guides, and deep dives into OpenClaw and AI agents.
Everything you need to know about MEMORY.md, session state, and context persistence in OpenClaw agents.
From Kraken connectors to BC integrations — the top skills for finance teams running OpenClaw.
Supply chain hardening, malicious pattern detection, and what’s coming in the security stack.
Step-by-step: use BC Connector + OpenClaw to automate journal entries and reconciliation.
Why autonomous finance agents are the next wave — and how to position yourself now.