Chris0x88

Onboarding Kit · Chris0x88

A generic AI-readiness audit you run yourself before booking a discovery call.

You point your AI agent at this folder. It interviews you, researches your business, captures your branding, detects which AI providers you already have configured, and produces a one-page HTML report that tells you exactly where AI fits into your operation. Run it in an afternoon. Bring the output to a discovery call and we start ten steps ahead.

The kit is harness-agnostic. It works with Claude Code, OpenAI Codex, Hermes, or any agent that can read markdown skill files.

What you walk away with

audit-report.html — your one-page client report. Open it in a browser. It covers what the agent learned about your business, three ranked AI integration opportunities specific to you, a recommended provider stack, and a 90-day picture of what good looks like.
consultant-brief.md — a structured markdown summary you can send to me, paste into an email, or drop into our first call. Same content as the client report but in a format I can read in 60 seconds.
branding.json — captured colours, fonts, logo URLs, voice samples. Useful for any future build work.
stack.json — what AI providers and tools the kit detected on your machine.

The kit does not phone home. Nothing is sent to me automatically. You decide what to share.

Five-minute install

git clone https://github.com/Chris0x88/Chris0x88.github.io.git
cd Chris0x88.github.io/consultant/onboarding-kit

Then attach the kit to your agent. See INSTALL.md for the exact invocation in:

Once attached, you say:

Run the onboarding kit.

The agent picks up kit.skill.md, runs the four sub-skills in order, asks you the interview questions, and writes the four output files into the working directory.

Total time: 30 to 60 minutes, depending on how thorough you want the business research pass to be.

What the kit actually does

Phase	Skill	What it produces
1. Interview	`interview.md`	Ten questions, captured verbatim
2. Research	`skills/business-research.md`	Public-facing summary of the company
3. Branding	`skills/brand-capture.md`	`branding.json` — colours, fonts, voice
4. Stack	`skills/stack-detection.md`	`stack.json` — providers and tools detected
5. Report	`skills/report-generator.md`	`audit-report.html` and `consultant-brief.md`

Each phase produces structured intermediate output the next phase consumes. The final report assembler ties it all together.

Privacy posture

Defaults are conservative.

No file content reading. The stack-detection skill checks for the presence of config directories and environment variables. It never opens or reads key values, never walks your file tree.
No browser history. No mail. No work directories beyond what you explicitly point the agent at.
The interview is voluntary at every step. Skip questions. Stop at any phase.
Nothing is transmitted off your machine unless you choose to send the consultant brief.

The full threat model and a list of every read the kit performs is in SECURITY.md.

When to skip the kit

You already have a clear, costed AI build brief. Send it. We’ll talk.
You’re not ready to run an agent on your own machine. That’s fine. Email me and we’ll book a discovery call without it.
Your data is genuinely too sensitive to involve a third-party model in any audit pass. Tell me up-front and we use a fully local stack.

Licence

MIT. Use it, fork it, customise it, run it for your own clients. Attribution appreciated.

If you build something good on top of this, I’d love to see it.

Book a discovery call · Field Notes · Resources