# CyperX.dev > Chris (CyperX) — builder, breaker, automator based in Brisbane, Australia. > AI agent infrastructure, distributed fleets, and tools that compound over time. CyperX.dev is a personal site and technical blog written for humans and agents alike. Content is semantic HTML with an RSS feed and machine-readable indexes. Each post is also available as raw markdown at /blog/.md (~5x cheaper to parse than the HTML page). ## Key pages - [Home](https://cyperx.dev/): overview and latest signals - [Blog](https://cyperx.dev/blog): long-form writing on agents, architecture, and building in public - [Projects](https://cyperx.dev/projects): things I build and maintain - [About](https://cyperx.dev/about): who I am and how I work - [Uses](https://cyperx.dev/uses): hardware, tools, and software I use daily ## Blog posts - [Delete-First: The Ablation Habit](https://cyperx.dev/blog/claude-code-ablation) (https://cyperx.dev/blog/claude-code-ablation.md): The people getting the least out of Claude Code have a 400-line CLAUDE.md and add a rule every time it slips. The people getting the most delete. Here's the ablation workflow, with a real before/after. - [Algebra for Agents: Orchestrating the Swarm](https://cyperx.dev/blog/claude-code-orchestration) (https://cyperx.dev/blog/claude-code-orchestration.md): Sequential, parallel, fan-out. A small set of operations that compose into work one agent can't do. Here's the real fan-out audit script, the referee that keeps it honest, and loops vs routines. - [Unhobbling: Goals, Not Steps](https://cyperx.dev/blog/claude-code-unhobbling) (https://cyperx.dev/blog/claude-code-unhobbling.md): You take a model that can plan, read a whole repo, and run your tests, and you cram it into a single-line edit. The horse can run — you tied its legs and you're proud it can still hop. Here's how to cut the rope. - [Verification: The One That Actually Matters](https://cyperx.dev/blog/claude-code-verification) (https://cyperx.dev/blog/claude-code-verification.md): Most prompt advice is cope. The single habit that moved my success rate more than any trick: give the model a mirror. Tests, types, screenshot diffs, and hooks that won't let it call broken code done. - [Anatomy of an Eve Agent: Five Files and Nothing Else](https://cyperx.dev/blog/eve-agents-anatomy) (https://cyperx.dev/blog/eve-agents-anatomy.md): An agent is a directory. A model file, a markdown loop, a typed schema, a folder of verbs, and a test suite that says whether it still works. Here's one taken apart, then the build order that starts with the eval. - [Eval-Gating: Don't Let an Agent Accept Its Own Output](https://cyperx.dev/blog/eve-agents-eval-gating) (https://cyperx.dev/blog/eve-agents-eval-gating.md): A judge subagent with zero tools, a zod schema the runtime enforces, and an if statement on a boolean. Generate, grade, then accept, retry, or park — the closed loop that stops an agent grading its own homework in prose. - [Eve Agents: Why Twelve Specialists Beat One God-Agent](https://cyperx.dev/blog/eve-agents-fleet-model) (https://cyperx.dev/blog/eve-agents-fleet-model.md): I built a fleet of twelve agents instead of one that does everything. Not because twelve is better at the work — because when one of them breaks, I know which one. Here's the roster and the reasoning. - [The Orchestrator Capstone: Composing Twelve Lessons Into One Agent](https://cyperx.dev/blog/eve-agents-orchestrator-capstone) (https://cyperx.dev/blog/eve-agents-orchestrator-capstone.md): Agent #12 introduces nothing new. It's forced fan-out plus eval-gating plus durable state, wired together — and a scorecard that scores all twelve at once. Including the row where the capstone is red. - [Root Lockdown: Delete the Tools and the Agent Has to Delegate](https://cyperx.dev/blog/eve-agents-root-lockdown) (https://cyperx.dev/blog/eve-agents-root-lockdown.md): My research agent ignored its own instructions and did 40 web searches itself instead of fanning out. The fix wasn't a better prompt — it was taking the search tools away. Structural constraints beat written ones. - [Using Claude Code Better: The Beginner's Guide](https://cyperx.dev/blog/using-claude-code-better-beginner) (https://cyperx.dev/blog/using-claude-code-better-beginner.md): Most people use Claude Code like fancy autocomplete and leave 90% of it on the table. It's a terminal agent that reads your repo, runs your commands, and checks its own work. Here's how to actually drive it. - [Using Claude Code Better: The Intermediate Guide](https://cyperx.dev/blog/using-claude-code-better-intermediate) (https://cyperx.dev/blog/using-claude-code-better-intermediate.md): You already live in the terminal with Claude Code. This is the gap between it works and it works like a senior engineer next to you: ablation, real permissions, verification hooks, and orchestration you can paste. ## Projects - [CYPERX.DEV](https://cyperx.dev/projects/cyperx-dev) — https://cyperx.dev: Personal site. Four themes. Zero restraint. - [CHANGELOGS.INFO](https://cyperx.dev/projects/changelogs-info) — https://changelogs.info: Release notes for tools that ship too fast to track. - [DOTFILES](https://cyperx.dev/projects/dotfiles): The machine as a living document. - [STOKEDAI](https://cyperx.dev/projects/stokedai): INTEL_PENDING — dossier being decrypted. ## Callable interfaces This site is callable, not just readable. Everything below is public, read-only, and needs no authentication. - MCP (preferred): https://cyperx.dev/mcp — Streamable HTTP. Tools: site_overview, search_posts, list_posts, get_post, list_projects, get_project. Card: https://cyperx.dev/.well-known/mcp/server-card.json - A2A: https://cyperx.dev/a2a — JSON-RPC message/send, answers inline. Card: https://cyperx.dev/.well-known/agent-card.json - API catalog (RFC 9727): https://cyperx.dev/.well-known/api-catalog - Agent Skills: https://cyperx.dev/.well-known/agent-skills/index.json — includes `query-cyperx-dev`, which documents all of the above in one file. Any content URL also answers in markdown when requested with `Accept: text/markdown` — no need to rewrite URLs. ## Machine-readable feeds - [RSS feed](https://cyperx.dev/rss.xml): all blog posts - [Sitemap](https://cyperx.dev/sitemap-index.xml): full URL index - [llms-full.txt](https://cyperx.dev/llms-full.txt): full text of site content in one file - [Content index](https://cyperx.dev/agents/index.json): structured JSON of every post and project ## Links - GitHub: https://github.com/cyperx84 - X / Twitter: https://x.com/cyperx