Integrator — I help people bring technology into their lives well, where it's genuinely needed and does the most good. Framework creator. Builder of tools that bridge human thinking and machine capability.
I've spent 30+ years in technology, and the thread through all of it is integration — taking complex things and making them work together. Based in Valencia, Spain, I build frameworks, MCP servers, and Claude Code tools that help people work with AI systematically rather than haphazardly.
I believe in integration over capability — that systematic implementation beats tool sophistication every time. The organizations that succeed with AI aren't the ones with the best tools; they're the ones with the clearest thinking about what they're trying to achieve.
16 npm packages · ~4,800 downloads/month · 72 Claude Code skills · 2 published books. Available for fractional AI & security advisory — the fastest way in is jimchristian.net.
Model Context Protocol servers connecting AI assistants to real-world services:
| Project | What It Does | |
|---|---|---|
| spain-ai-kit 🇪🇸 | Monorepo of MCP servers for Spanish government open data — INE statistics, BOE legislation (12k+ laws), Catastro land registry, AEMET weather |  |
| mcp-arr | Media management suite (Sonarr, Radarr, Lidarr, Readarr, Prowlarr) |  |
| mcp-tailscale | Tailscale network management |  |
| mcp-local-llm | Route mechanical tasks to local LLMs via Ollama |  |
| mcp-wisdom | Philosophy-grounded thinking frameworks (Stoic, Cognitive, Strategic) |  |
| mcp-threatintel | Unified threat intelligence (OTX, AbuseIPDB, GreyNoise, abuse.ch) |  |
| mcp-cti-glossary | Cyber threat-intel glossary — MITRE ATT&CK, OFAC SDN, and a cross-vendor threat-actor alias map with full-text search |  |
| mcp-fantastical | Fantastical calendar management |  |
| mcp-pihole | Pi-hole v6 DNS blocking control |  |
| mcp-homescan | Home network discovery and security scanning |  |
Skills, agents, and tools for extending Claude Code:
| Project | What It Does | |
|---|---|---|
| daily-patterns-pack | Session logging + Obsidian integration + pattern analysis |  |
| claude-agent-borg | Analyze external Claude setups, assimilate the best features |  |
| claude-lessons-learned | Structured retrospective analysis from incidents |  |
| cheaters-generator | Personalized cheatsheet systems |  |
| the-antislop | Detect and fix AI-generated writing patterns (35+ patterns) |  |
| claude-ecosystem-health | Drift detection across skills, agents, MCP servers, and vault |  |
| claude-draft-reviewer | Review and fix drafts for slop, craft, and voice consistency |  |
| claude-rfu-audit | 11-gate utility gauntlet — validate project value before investing effort |  |
| claude-art-skill | Visual content system — 16 workflows, 2 AI models, aesthetic routing, brand customization |  |
| signal-starter-kit | Four Claude Code skills (think-first, antislop, voice-editor, lessons-learned) with a one-command installer |  |
csAIi is my current framework — Context, Skills, Agents, Integration, Iteration: a discipline for working with AI as a co-operating system rather than a transactional tool. Explore it as an interactive walkthrough.
I've also open-sourced two complementary methodologies for AI adoption and implementation under CC BY-SA 4.0:
| Framework | What It Does | |
|---|---|---|
| The PAST Framework | Strategic clarity for AI adoption — Purpose, Audience, Scope, Tone |  |
| The SHAPE Methodology | Execution methodology — Situation, Hypothesis, Action, Process, Evaluation |  |
PAST tells you what and why. SHAPE tells you how and when. Together they take AI initiatives from scattered experiments to systematic implementation.
| Project | What It Does | |
|---|---|---|
| minervia-starter-kit | A co-operating system for human-led knowledge work |  |
| wisdom-pack | Philosophy-grounded slash commands for Claude Code |  |
| book-indexer | Index books and PDFs into ChromaDB for semantic search |  |
| cfbr | CLI wrapper around Cloudflare Browser Rendering — fetch JS-rendered pages as Markdown, JSON, screenshots, or PDFs |  |
| dayone-to-obsidian | Export Day One journal entries into Obsidian daily notes |  |
I publish Signal Over Noise — a newsletter about getting real value from AI without the hype. Evidence-based, practically skeptical, no AI slop allowed.
I'm also the author of How to Think Like a Coder (Batsford, 2017) — computational thinking taught without a line of code, shortlisted for the ALCS Educational Writers' Award — and How to Code in Minecraft (Dennis Publishing, 2016).
