Tools.

Things you can run, install, or import: a live RAG chatbot, three open-source kits of Agent Skills for Claude and ChatGPT, and n8n workflows for content pipelines. These are not demos. They're the same artifacts I use day to day, packaged so you can take them.

Governance documents (policies, risk scoring, measurement, vendor evaluation) are on /frameworks. The per-skill directory, with a link to every SKILL.md, is on /skills. If something here is broken or unclear, tell me and I'll fix it.


Hosted and running right now. Nothing to install.

Marketing Truths RAG rag chatbot

Ask a marketing strategy question in plain English and get an answer grounded in peer-reviewed science, with the sources attached. It retrieves against 200+ empirical generalizations distilled from more than 10,000 peer-reviewed studies across 90 marketing capabilities, so the answer cites evidence instead of asserting best practice.

Co-built with the Marketing Science Institute.

ask it a question ↗

Three public repos, MIT. Two are packs of Agent Skills, the open standard Anthropic published and OpenAI adopted, so one SKILL.md folder installs unchanged in Claude Code, claude.ai, ChatGPT, and Codex. The third is a Claude Code template. For every skill listed individually, with a link to each file, see /skills.


15 importable workflow JSON files. Drop one into your n8n instance, swap the placeholder credentials for your own, ship. Every real ID is scrubbed before publication. They live in alectivism/n8n-workflows, MIT-licensed.


Need a tool that doesn't exist yet?

If you're shipping an agent system and there's an obvious tool-shaped hole (an eval pattern, a governance hook, a workflow I haven't published), tell me what's missing. If it generalizes, I'll build and open-source it.

request a tool →