Resources: tools and artefacts
Skills, tools and interactive artefacts I have made that you can install, adapt or explore.
Most of what I publish here is writing. Resources is where I collect the things made to be used as well as read: practical skills for AI agents, visual ways of working and small interactive artefacts.
Use them as they are, or adapt them to your own work. I typically publish the method as well as the output so that other people can try it, question it, change it and use it to improve both the work and its outcomes.
Featured
Product Idea Pack
Turn a rough digital-product idea into something people can see, discuss and change before anyone commits to polished UI.
Product Idea Pack is an open Agent Skill for creating proposition sketches, mechanism boards, product storyboards and interactive prototypes. Start from a short prompt, a developed conversation or a guided intake; use it alone or in a workshop.
Outputs remain editable, with optional PowerPoint, PDF and Figma hand-offs. In the release tests, a typical generation or revision turn took about 15–20 minutes. Inspect and refine the result for its intended audience.
Read how and why I made it: Making product ideas visible before they look finished.

Agent skills
An Agent Skill is a folder of plain-language instructions and supporting files that teaches a compatible AI agent how to carry out a focused workflow. You can read one before installing it, and because the format is open you can change it to fit your own practice. These skills can be used in ChatGPT/Codex, Claude Cowork, Claude Code and other compatible Agent Skills harnesses; the available tools affect which parts of a workflow can be completed directly.
Using rafeblandford.com
Choose the right machine interface for this site—MCP, the public Content API or plain Markdown—and avoid overstating what a selected public record proves.
Read the skill · Install from GitHub
Verifying AI crawlers
Separate genuine AI crawler traffic from requests that merely claim a familiar user-agent, using signatures or first-party IP ranges and appropriately cautious reporting.
Read the skill · Install from GitHub
The complete public catalogue, installation routes and licence are at github.com/rafeblandford/skills.
Interactive artefacts
Small, self-contained diagrams and explorers that are easier to understand by using than by describing:
- Agentic product loops — how AI changes the economics of feedback loops in product development.
- RafeOS architecture — how the personal system I write about fits together.
Explore the full artefact gallery.
For machines
Published skills are indexed at /.well-known/agent-skills/index.json, with a SHA-256 digest for each entry file. The same resources are represented in the site's ARD and compatibility AI Catalog documents. Other machine interfaces are described in the colophon and /llms.txt.