zurbrick

ACTIVATE THIS SKILL when you make a mistake, receive a correction, discover something new, or are about to start major work. Captures errors, corrections, and discoveries in a fast one-line format, deduplicates them, queues recurring or critical lessons for human approval, and scans past failures before major tasks to prevent repeats. Use it every time something goes wrong — no exceptions.

19 production-hardened patterns for AI agents — memory, verification, ambiguity handling, compaction survival, delegation, proof-based handoffs, stale-worker recovery, and self-improvement. Use when hardening an agent for production reliability, when an agent keeps hallucinating or losing context, when handoffs between agents drop details, when delegated work silently fails, or when someone says "my agent is unreliable" or "how do I make this more robust." Works with OpenClaw, Claude Code, Cowork, or any SKILL.md-based agent setup. Includes the Isolated Agent Fabrication Guard plus new delegation hardening patterns for brief quality, completion contracts, acceptance gates, silent-worker recovery, and scoped verifier use.

Route high-value content into a Notion workspace with a quality gate, destination mapping, and exact MCP write patterns. ALWAYS trigger when the user says "save this to Notion", "put this in my brain", "add this to the knowledge base", "capture this for later", "log this in Notion", or any variation of saving, storing, or routing content to Notion. Also trigger when an agent produces a research summary, decision memo, project plan, status update, article draft, security audit, financial snapshot, weekly rollup, contact note, meeting prep, or other durable content that belongs in Notion. Trigger on mentions of a knowledge hub, Inbox DB, content hub, work hub, or finance hub as destinations. Skip for health workflows, property/equipment tracking, or workspace memory-only writes.

Lock down any LLM agent against prompt injection, data exfiltration, social engineering, and channel-based attacks. Use when setting up a new agent, auditing an existing agent's security posture, hardening an agent that handles sensitive data, reviewing MCP server permissions, or when someone says "how do I make this agent more secure" or "protect against prompt injection." Works with OpenClaw, Claude Code, LangChain, custom MCP setups, and any agent framework that accepts natural-language input and calls external tools.

Audit and improve an agent at the right layer: persona/tone, constitutional and operating rules, memory architecture, or skill portfolio / reusable workflows. Use when an agent feels off, brittle, repetitive, forgetful, over-scoped, or uneven and the question is which lane is actually weak and what the smallest justified fix is. Helps decide whether to make a plain edit, tighten a rule, adjust memory, change a skill, or do nothing.

Guide for OpenClaw setup, config, commands, routing, and troubleshooting. Use when the user asks how OpenClaw works, how to configure it, why a channel or agent is misbehaving, or how to diagnose gateway/channel/session issues. Prefer local docs first, then inspect the specific config subtree before suggesting changes. Not for designing or auditing skills themselves.

Encrypted backup and restore for OpenClaw agents. Creates two-tier archives: cloud-safe operational backups plus optional age-encrypted secrets for local recovery. Use when setting up disaster recovery, backing up your agent, verifying backup integrity, restoring from backup, or pushing safe archives to GitHub.

Practical memory architecture for OpenClaw agents. Use when an agent needs to remember commitments across sessions, retrieve prior context, prevent repeat mistakes, stage durable-memory updates safely, compile reusable procedures, or improve memory hygiene over time. Supports a 3-tier adoption path: Core, Recommended, and Advanced.

Troubleshoot, diagnose, and optimize OpenClaw deployments. Use this skill whenever the user mentions OpenClaw issues, gateway problems, stuck agents, channel failures (Telegram/Discord/iMessage/Slack), session locks, cron job failures, heartbeat issues, or message delivery problems. Triggers on "openclaw doctor", "gateway not responding", "bot not replying", "agent stuck", "messages not delivering", "Telegram bot broken", "openclaw won't start", "session stuck", "cron not running", "heartbeat failed", "channel warnings", "openclaw update", "openclaw config", "my bot is down", or "the agent isn't working".

Fast OODA-loop decision skill for high-uncertainty situations. Use when the user wants to work through Observe → Orient → Decide → Act, make a call quickly, pressure-test a situation, or determine the next move under uncertainty.

Output validation gates for AI agent systems. Prevents hallucinated data, leaked internal context, wrong formats, duplicate sends, post-compaction drift, and false delegated completions. Use when building or operating an agent that delivers output to humans or external systems. Provides a tiered gate system (internal → user-facing → external → code), protocol gates for recurring failure modes, delegated-work acceptance gates, severity classification, and a feedback loop for gate evolution. Triggers on phrases like "QA gates", "validation", "output quality", "prevent hallucination", "delivery checklist", "agent QA".

Generate an interactive HTML observability dashboard for your OpenClaw deployment. Shows agent health, channel status, session activity, token usage, cron jobs, and error trends — all in one visual view. Use this skill whenever the user says "how are my agents doing", "show me agent status", "openclaw dashboard", "observability", "monitoring dashboard", "are my channels healthy", "show session activity", "token usage report", "agent health check", "what's the state of my openclaw", or any request for a visual overview of their OpenClaw deployment. Also triggers on "generate a status report for my agents", "which agents are active", "session count", or "how much am I spending on tokens". Even casual requests like "how's openclaw looking" or "give me the rundown on my agents" should trigger this skill.