Agents Guides
54 InsiderLLM guides in Agents — practical, tested walkthroughs for running AI locally, sorted by most recently updated.
- Running OpenClaw 100% Local — Zero API Costs Configure OpenClaw to run entirely through Ollama with no API keys, no cloud calls, and no monthly bills. Full setup guide with model picks by VRAM tier.
- Best Hardware for Running OpenClaw — Mac Mini vs VPS vs Your Old PC OpenClaw runs 24/7. A Mac Mini M4 draws 4 watts idle. A free Oracle VPS costs nothing. A used ThinkCentre costs $85. Here's which one to pick.
- Slash Your AI Costs With a Token Audit Your AI API bill is higher than it needs to be. A 15-minute token audit finds the waste — system prompts, ballooning history, hidden tool tokens. Here's the exact process.
- Anthropic Just Cut Off OpenClaw Users — Why Local Models Matter More Than Ever Starting April 4, Claude subscribers can no longer use their subscription for OpenClaw and other third-party harnesses. If you were relying on cloud AI for your agent, here's how to go fully local.
- Jev Mode on a 3090: 26 ms per Token You Don't Write Jev-style parallel decisions in llama.cpp on Qwen3.6-27B, RTX 3090: 1.3x faster at four tokens, 5.5x at fifty, 21 vs 23 of 47. The gain is the tokens you skip.
- What Is Jev, and Can You Run One on Your Own GPU? Jev picks from options you supply instead of writing, in one pass, with a probability per answer. What it is, what it costs, and the open version for a 3090.
- LoRA Skill Compilation Is a Double-Headed Coin Flip Ten LoRA seeds on identical data spread 3.62 points, against 4.65 for prompt compilation. Not one beat the no-adapter base. Measured on an RTX 3090.
- I Got the AI Result I Wanted. Then I Ran It Nine More Times A frontier model read my logs and wrote a skill that beat my local 27B baseline by 10.6 points. Nine more compilation runs showed the number was fake.
- Skills in the Weights: The LoRA Answer to the Prompt Tax Compiling a skill into a prompt costs 1,383.9 tokens per call, forever. Putting it in a LoRA costs 7.55 GB and a training run. Only one has been measured.
- epsiclaw: OpenClaw Stripped to 515 Lines of Python (The Karpathy Treatment) epsiclaw is a minimal, readable reimplementation of OpenClaw in 515 lines of Python with 6 files and one dependency. Inspired by Karpathy's approach to autoresearch. Here's what it does and why it matters.
- OpenClaw vs Cursor: Local AI Agent or Cloud IDE? OpenClaw is free, private, and runs your own models. Cursor is polished, fast, and cloud-powered. A developer's comparison: cost, privacy, model flexibility, offline use, and where each one wins.
- Claude Code vs PI Agent — Which Coding Agent for Local AI? Claude Code vs PI Agent compared for local AI development. System prompts, tools, pricing, local model support, and honest verdicts for every type of developer.
- Best Local LLMs for Function Calling: Qwen 3.6, Gemma 4 Function calling with local LLMs on Ollama and llama.cpp. Current lineup: Qwen 3.6, Gemma 4, DeepSeek V4. Common failures, agentic loop patterns. July 2026.
- Best Local Alternatives to Claude Code in 2026 Aider, Continue.dev, Cline, OpenCode, Void, and Tabby compared. Which open-source coding tools work best with local models on your own GPU?
- Qwen 3.6: Why Q4 Quant Breaks Local Coding Agents (And the Fix) A viral thread says Q4-to-Q6 fixes Qwen 3.6 coding, but the test was confounded. What four independent reports show about the quant tax on coding agents.
- Best Local Models for OpenClaw 2026: Qwen 3.6 + DeepSeek V4 Qwen 3.6-27B dense ties Sonnet 4.6 on agentic coding; 3.6-35B-A3B runs OpenClaw on 16GB VRAM. Plus DeepSeek V4-Flash, sampling tips, VRAM tiers.
- Best Local Models for PI Agent: Qwen 3.6, Gemma 4 (2026 Setup) PI Agent runs any model locally via Ollama. May 2026 picks: Qwen 3.6 27B / 35B-A3B MoE, Gemma 4 26B-A4B. Setup, model comparisons, honest limits.
- Best OpenClaw Plugins and Skills Guide (2026) Every OpenClaw skill worth installing, how to avoid malicious plugins on ClawHub, and how to build your own. 1,103 of 14,706 skills are malicious.
- Best Ways to Fix OpenClaw Tool Call Failures: 2026 Guide Your OpenClaw agent silently fails, loops, or corrupts its session. Six debug paths plus May 2026 gotchas: Qwen 3.6 whitespace kwargs, Gemma 4 thinking mode.
- Best Local LLMs for Structured Output: Qwen 3.6, Gemma 4 JSON schema, grammar constraints, and Outlines compared. Current model picks: Qwen 3.6, Gemma 4, DeepSeek V4. Common failures + working code. May 2026.
- OpenClaw Memory Problems: Context Rot and the Forgetting Fix Your OpenClaw agent forgets instructions, repeats questions, contradicts itself. How memory and context rot work — and fixes that apply to any harness.
- OpenClaw Security Guide: Risks and Hardening 42,000+ exposed instances, Google suspending accounts that connected via OAuth, 26% of ClawHub skills with vulnerabilities. Real risks, prompt injection demos, and step-by-step hardening for OpenClaw.
- How OpenClaw Actually Works: Architecture Guide 5 input types explain the 'alive' behavior: messages, heartbeats, crons, hooks, and webhooks feed a single agent loop. The 3am phone call was just a timer event.
- Fix OpenClaw Token Waste: $150 to $6 Overnight Cut OpenClaw API costs by 97% with three proven fixes: route heartbeats through Ollama, add tiered model routing, and purge session history token bloat.
- Best Way to Set Up OpenClaw (2026 Guide) Run `npx openclaw@latest`, scan a QR code for WhatsApp, and your AI agent is live. Gateway needs just 2-4GB RAM. Add Ollama for local models or connect Claude/GPT-4 via API.
- OpenClaw ClawHub Alert: 1,103 Malicious Skills Found OpenClaw ClawHub security alert: 1,103 malicious skills found across 14,706 audited. CVE-2026-28458 Browser Relay auth bypass. How to protect yourself now.
- Best OpenClaw Alternatives: 11 Tools That Actually Work in 2026 Tested alternatives to OpenClaw for local AI agent workflows — Hermes Agent, Pi Agent, MMX-CLI, VT Code, and seven more. Ranked by setup ease, model support, and what actually works after Anthropic's April 4 subscription cutoff.
- OpenClaw Critical Sandbox Escape: Update to 2026.3.28 Now Ant AI Security Lab found 33 vulnerabilities in OpenClaw including critical privilege escalation and filesystem sandbox escape. If you're self-hosting, update immediately.
- 12 Architecture Patterns from the Claude Code Leak -- Ranked by Payoff for Local AI Claude Code's leaked source reveals 12 engineering patterns that power a $2.5B product. Ranked by how much each one improves your local AI agent setup.
- Claude Code's Source Just Leaked: What 500K Lines of TypeScript Reveal About AI Coding Agents Claude Code's full source was exposed via npm source maps. Here's what the leaked architecture reveals about multi-agent orchestration, and what it means for local AI agent builders.
- OpenClaw Trading Scams: How to Spot AI Agent Grifts Before They Cost You AI agent trading scams target technical users who know agent pipelines are buildable. Here's the playbook they use, the math that breaks their claims, and how to protect yourself.
- How to Run Karpathy's Autoresearch on Your Local GPU Set up Karpathy's autoresearch on your GPU to run 100+ ML experiments overnight. Works on RTX 3090/4090 as-is, scales down to 6GB cards with tweaks.
- Best Ways to Connect Local AI to Notion in 2026 4 real ways to connect Notion to a local LLM without sending data to the cloud. MCP servers, RAG pipelines, Open WebUI, and n8n workflows compared with setup steps.
- OpenClaw Security Report: January 2026 Three high-severity CVEs, a supply chain attack on ClawHub, and 21,000+ exposed instances. Every OpenClaw security event from January 2026 with sources.
- Best OpenClaw Tools and Extensions in 2026 Crabwalk visualizes agent actions in real time, Tokscale catches API bills before they hit $200+, and openclaw-docker locks down deployment. The best 3rd-party tools ranked.
- OpenClaw Security Report: February 2026 — ClawHub Malware, Google Suspensions, and Critical Fixes 17 security fixes, 341 malicious ClawHub skills, Google banning users, and the creator leaving. Every OpenClaw security event from February 2026.
- OpenClaw on Raspberry Pi: What Actually Works (and What Doesn't) Pi 5 with 8GB RAM runs OpenClaw as a gateway with cloud APIs. Local LLMs hit 2-7 tok/s on 1.5B-3B models. Step-by-step setup for llama.cpp, Ollama, and OpenClaw on ARM64.
- OpenClaw Model Combinations: What to Pair for Each Task Stop running one model for everything in OpenClaw. Pair Qwen 2.5 Coder 32B for autocomplete, Qwen 3.5 27B for planning, and Qwen3-Coder-Next for agentic coding. Combos by VRAM tier.
- OpenClaw Security Hardening — Every Fix in February 2026 SSRF bypass, sandbox escapes, unauthorized disk writes, session hijacking. Every security fix OpenClaw shipped in February 2026, explained in plain English.
- OpenClaw on Mac: Setup, Optimization, and What Actually Works brew install openclaw-cli, connect Ollama, configure the gateway, and stop fighting macOS. Apple Silicon setup, memory math, launchd config, and the gotchas nobody warns you about.
- OpenClaw After Steinberger — What the OpenAI Move Means for Your Setup Peter Steinberger joined OpenAI. Three releases shipped since. Elon Musk posted a meme. Baby Keem is debugging agents. Here's what actually matters for your OpenClaw setup.
- Prompt Debt: When Your System Prompt Becomes Unmaintainable Spaghetti Your system prompt started at 200 words. Six months later it's 3,000 words of contradictory instructions and panic patches. Here's how prompt debt accumulates, what it costs, and how to pay it down.
- LightClaw: A 7,000-Line Python Alternative to OpenClaw OpenClaw is 40,000+ lines of TypeScript. LightClaw does Telegram AI assistant, 6 LLM providers, memory, skills, and agent delegation in ~7,000 lines of Python. One week old, 12 stars, one developer. Here's what it can and can't do.
- Building AI Agents with Local LLMs: A Practical Guide Build AI agents with local LLMs using Ollama and Python. Model requirements, VRAM budgets, framework comparison, working code example, and security warnings.
- The Web Is Forking: What the Agentic Web Means for Local AI Builders Coinbase, Stripe, Cloudflare, Google, OpenAI, and Visa are building a parallel web for AI agents. Money, search, content, execution — all redesigned for software clients. What local AI builders should do now.
- SmarterRouter: A VRAM-Aware LLM Gateway for Your Local AI Lab Intelligent router that profiles your models, manages VRAM, caches responses semantically, and auto-picks the best model per prompt. Works with Ollama and llama.cpp.
- LocalAgent: A Local-First Agent Runtime That Actually Cares About Safety Rust CLI for AI agents with deny-by-default permissions, approval workflows, and deterministic replay. Works with LM Studio, Ollama, and llama.cpp.
- The 5 Levels of AI Coding: Where Are You, and Where Is This Going? A 3-person team ships production Rust with zero human code. Most devs using AI get 19% slower. The gap between these facts is where software development lives now.
- OpenClaw's Creator Just Joined OpenAI — Here's What It Means for Local AI Agents Peter Steinberger built the fastest-growing open-source project ever. Now he's at OpenAI. OpenClaw stays open. Here's what changes for local AI builders.
- OpenClaw Model Routing: Cheap Models for Simple Tasks, Smart Models When Needed Stop paying Opus prices for heartbeats. Set up tiered model routing in openclaw.json so cheap models handle 80% of work and frontier models only fire when needed.
- What Agents Can't Do (Yet): The Seven Human Capabilities Missing from AI Systems SOUL.md files are bandaids. Agents are getting smarter but not wiser — intelligence without restraint. Seven capabilities humans use instinctively that no agent framework has solved, and a gate-based architecture that might.
- ClawHub Malware Alert: Top Skills Infected The #1 ClawHub skill was malware. How it stole API keys, Cisco's new scanner tool, the MoldBot data leak, and 7 things to do right now to protect yourself.
- Stop Using Frontier AI for Everything Build a tiered AI model strategy that stops wasting money on GPT-4 and Claude Opus. Route tasks to local models, Haiku, Sonnet, or Opus based on complexity.
- OpenClaw vs Commercial AI Agents: Which Should You Use? OpenClaw costs $0 plus API fees. Lindy runs $49-299/month but has 7,000+ integrations and SOC 2 compliance. Privacy, reliability, and customization compared honestly.
A weekly email with every new guide and measured benchmark.
Subscribe — free, no spam