ai building
Claude Code AGENTS.md Support: When CLAUDE.md Still Wins
Claude Code reads AGENTS.md since 2.1.277, but only when the project has no CLAUDE.md. If both exist, CLAUDE.md wins. The four modes, the one-line fix and the limits.
· 8 min
#ai-agents
10 posts tagged with "ai-agents"
adhdaeoagent-architectureagent-harnessagent-readinessagent-skillsagent-telemetryagentic-commerce-protocolagentic-seoagentic-systemsagentsagents-mdaiai-agentsai-benchmarksai-buildingai-coding-toolsai-crawlersai-engineeringai-searchai-toolsai-visibilityai-workflowsanalysisanswer-engine-optimizationanthropicapiarchitectureautomationavr-frameworkbinancebrowser-aibrowser-ai-apibug-bountybuilderbusinesschatgptchrome-prompt-apicitation-mechanicsclaudeclaude-adhdclaude-apiclaude-codeclaude-fable-5claude-fable-5-1claude-fable-5-vs-opus-4-8claude-fable-5-vs-opus-5claude-maxclaude-max-planclaude-nerfedclaude-opus-4-8claude-opus-5claude-opus-5-5claude-pricingclaude-skillsclaude-usage-limitsclaudebotclient-side-aicloudflarecodexcomparisoncontent-creationcontent-intentcontent-optimizationcontent-strategycontextcross-postingcursordebuggingdeploymentdeveloper-productivitydeveloper-toolsdevopsdevtodigitaloceandistributed-systemsdomain-authoritydopamine-menuentity-graphentity-seoevaluationexecutive-functionfable-5feesfuturegemini-nanogeogithubgithub-copilotharness-engineeringhookshostingimplementationinfrastructureis-claude-nerfedkalshikelly-criterionknowledge-graphllmllms.txtmathmcpmethodmicrosaasmodel-context-protocolmodel-routingmulti-agentnavigator-modelcontextneurodivergentnodejsnotionopenclawoptimizationopus-5orchestrationoriginalitypattern-recognitionpdfperplexityphilosophyplan-quotapluginspolymarketposition-sizingprediction-marketspricingproductionproductivityprompt-cachingpromptspythonqualityquality-controlragrailwayresearchroadmaprobots-txtrobots.txtsaassameasschemaschema-orgsearch-consolesearch-enginessecurityseoserverlessshared-payment-tokensskillssoftware-developmentsoftware-engineeringstartupstrategystripe-link-walletstructured-datasubagentssveltesveltekitsystems-designtokenstradingtrading-botstutorialvercelverificationvpsweb-standardswebmcpwebsocketworkflowzapierzero-cost-ai-audit
ai building
Claude Code Skills vs Subagents vs Hooks vs CLAUDE.md: Which One Fits Which Job
Claude Code skills vs subagents vs hooks vs CLAUDE.md as of 09/29/2026: when each loads, whose context it uses, and a simple decision sequence for picking one.
· 12 min
ai building
Claude Code Skills: What They Are, Where They Live, and How to Write One
Claude Code skills explained: what a SKILL.md is, where skills live, how to write one, and the skills I actually run. Checked against the docs on 09/29/2026.
· 6 min
ai building
SvelteKit MCP: The Official Svelte Server, and What Its Autofixer Found in 134 Components
The official Svelte MCP server gives Claude Code current Svelte 5 docs and an autofixer. I ran it on 134 components: 70 missing keys, 34 files it could not read.
· 7 min
ai building
Fable 5.1 vs Opus 5 in Production: A Whole Day on Fable Used 19% of a Max Weekly Window
Same harness, same day: 2,443 Fable 5.1 turns and 1,120 Opus 5 turns after a weekly reset consumed 19% of the Max weekly window, and the 5-hour window peaked at 59% without capping. Fable errored less per tool call, hit more permission gates, and would have cost 1.4x per turn on the API, not 2x.
· 15 min
automation
Run Your AI Agent on a VPS: The $6/Month Always-On Setup
Your AI agent dies when your laptop sleeps. The $6-25/month VPS setup I use to run trading agents 24/7: specs, systemd, and when latency justifies more.
· 6 min
ai building
How to Use Claude Opus 5: A Failure-Tested Guide
Learn how to use Claude Opus 5 with failure replay, effort sweeps, deterministic checks, and a scheduled regression worker.
· 11 min
ai building
What Is WebMCP?
WebMCP is an experimental browser API that lets websites expose structured tools to AI agents through document.modelContext. Learn how it works, its security boundaries, browser support, and how it differs from MCP.
· 7 min
ai building
Claude Fable 5 System Card, Annotated: 95% Capability, Documented Fabrication
The Fable 5 system card documents both halves: 95.0% SWE-bench Verified, and fabricated status reports with unverbalized early-stops. The annotated read.
· 7 min
ai building
I Built a Private MCP Server to Give Claude Memory Across Sessions. Here Is What Broke.
I shipped a private MCP server bridging my knowledge base into claude.ai via OAuth 2.1: the architecture, two bugs the smoke test missed, and the isolation pattern.
· 8 min