OpenClaw Guides

Every published guide for OpenClaw operators, organized by problem type. Pick the one that sounds like yours.

The Money Problem

API spend higher than expected? These fix it.

Spend limits, model selection, prompt caching, rate limits.  See hub page →

Setting spend limits so your agent stops at night
Structuring your system prompt for maximum cache hits
OpenClaw uses your most expensive model for everything, even simple tasks
Choosing a model based on your actual workload
How much does it actually cost to run OpenClaw for a month?
I turned on prompt caching and my bill didn’t change at all
I woke up to a $300 OpenClaw bill and had no idea what caused it
OpenClaw keeps hitting rate limits every day
OpenClaw keeps switching models mid-task and I don’t know why
OpenClaw is ignoring my model override
OpenClaw responses on Telegram are too slow
How to test whether a local model can handle your workload before committing
I switched to a cheaper model and my agent got worse
My OpenClaw responses got noticeably worse after I switched to a cheaper model
Which OpenClaw features cost money and which ones are completely free

The Setup Problem

Installing, configuring, or recovering from a broken instance.

Config, integrations, plugins, recovery, debugging, performance.  See hub page →

I just installed OpenClaw. What do I actually need to configure before using it.
The OpenClaw config fields that look optional but will cause problems if you skip them
OpenClaw won’t start after I changed the config
How to roll back an OpenClaw config change when you can’t get back in
How to update OpenClaw without losing your config
How to move OpenClaw to a new server without losing anything
How to run OpenClaw on a $5 VPS without it falling over
OpenClaw plugins vs skills: what the difference actually means
A plugin I installed broke OpenClaw. How to recover.
How to connect OpenClaw to Telegram without exposing your bot token
How to connect OpenClaw to Discord so it responds in your server
How to connect OpenClaw to WhatsApp
OpenClaw Telegram integration isn’t working and I don’t know why
How to route different tasks to different messaging channels in OpenClaw
How to reduce system prompt size without losing context
Why is OpenClaw so slow? It’s probably your context window.
Why does OpenClaw fill up context so fast even on simple tasks?
Why does OpenClaw keep compacting even on short sessions?
My OpenClaw gateway keeps restarting on its own
OpenClaw started in a broken state and now nothing works
My OpenClaw agent stopped being able to run commands after an update
OpenClaw tool calls are failing silently
OpenClaw is making API calls I never asked for. Here’s what’s causing them.
My OpenClaw agent is looping on the same task over and over
OpenClaw created a second instance I didn’t ask for
My OpenClaw skill isn’t loading
How to reproduce an OpenClaw bug before you report it

The Security Problem

Exposed instances, access control, vulnerabilities, credential leaks.

Gateway exposure, permissions hardening, CVEs, incident response, plugin safety.  See hub page →

The Memory Problem

Your agent forgets. Here’s how to fix that for good.

Memory plugins, embedding models, recall quality, compaction, scopes.  See hub page →

The Autonomy Problem

Running tasks while you’re away, without babysitting.

Cron jobs, task queues, sub-agents, notifications, workflows.  See hub page →

Cron scheduling patterns that work
My OpenClaw cron job ran twice — or never ran at all
How to stop OpenClaw cron jobs from piling up when tasks run long
My OpenClaw cron job works in testing but silently does nothing in production
How to schedule a daily task in OpenClaw without building a queue system
Task B ran before Task A finished and everything broke
How to pass output from one OpenClaw cron run into the next
My OpenClaw agent failed overnight and I didn’t find out until morning
My cron job ran but the Telegram notification never arrived
How to know what your agent actually did while you were away
My OpenClaw agent won’t spawn sub-agents
OpenClaw keeps asking me to approve sub-agent device pairing and I can’t automate past it
OpenClaw sub-agents can’t access the skills I installed on my main agent
How to run two OpenClaw agents on one server without them conflicting
How to route different people’s messages to different OpenClaw agents
How to build a morning brief with OpenClaw that takes 30 seconds to read
How to build a weekly digest with OpenClaw that reads like a human wrote it