Posts
OpenClaw Operations - Building the Flywheel
· β˜• 3 min read
1. Production Cases Cleaning up a node Restarting a node Deleting an abnormal Pod Inspecting the cluster 2. Two Steps to Get Configured Step one: configure MCP. Hand this configuration to OpenClaw and it will set it up for you. 1 2 3 4 5 6 7 8 { "mcpServers": { "ops-mcp-server": { "baseUrl": "http://ops-mcp-server/mcp", "headers": { "Authorization": "Bearer xxx" } } } } Step two: install skills.

You Can Also Implement AI Agents This Way - The SOPS Edition
· β˜• 5 min read
1. Establish a SOPS Baseline SOPS stands for Standard Operating Procedures. In the production activities we carry out, there are often some basic, repetitive operational flows, and these operations are the SOPS. You need to build up a set of SOPS; by combining them you can finish tasks faster. In the past, establishing standard procedures required a senior engineer, but now with AI, producing SOPS is very easy.

AGFS File System Performance Test
· β˜• 6 min read
1. Installing wrk 1 apt-get install -y wrk 2. localfs Performance Test 2.1 1KB Small File Test Write file 1 2 3 4 5 6 7 8 9 10 11 12 13 14 cat > put-1k.lua <<"EOF" wrk.method = "PUT" wrk.body = string.rep("a", 1024) wrk.headers["Content-Type"] = "text/plain" counter = 0 local WRAP = 500000 request = function() counter = counter + 1 local n = ((counter - 1) % WRAP) + 1 local filename = "/local/perf-1k/test_" .

A File System for AI Agents - AGFS
· β˜• 5 min read
1. Introduction to AGFS Overview AGFS is a project that provides storage services through a RESTful API. It supports a variety of storage backends, including memory, files, databases, message queues, and more. Because it does not need POSIX interface support, data can be stored through remote calls, which is very useful in AI Agents.

OpenViking Usage and OpenClaw Integration
· β˜• 3 min read
1. Introduction to OpenViking 1.1 Use Cases Problem How OpenViking Handles It Fragmented context Unified management with a filesystem paradigm: memory, resources, and skills all map to a viking:// virtual directory, so context is managed the way files are High token consumption L0/L1/L2 tiering: the L0 summary (~100 tokens) is