LCF Consulting · Developer tooling
Drop-in guardrails for Claude Code, Codex & Cursor agents
A practical companion to Harness Engineering for AI Coding Agents. Configure agent instructions, safer tool use, and CI checks without starting from scratch.
Inside the kit
- Three pairs of AGENTS.md and CLAUDE.md templates: generic, Python data pipeline, TypeScript web
- Claude Code permission settings and five guardrail hooks
- CI gates for tests, lint, secret scanning and diff size
- Task specification, context pack and evaluation checklists
- 12-page Quickstart PDF and usage license
Personal
Use in your own projects. No resale or redistribution.
Team
Use by up to 10 developers within the same organization. No redistribution.
Frequently asked questions
How is it delivered?
After completing payment, use PayPal’s return-to-merchant link if you are not redirected automatically. The delivery page starts your ZIP download and also provides a download button. Keep your delivery link private.
Does this work with my agent?
Instructions and CI templates are portable. Claude-specific hooks/settings require adaptation for other tools.
What is the refund policy?
Contact us within 14 days if the purchased files cannot be delivered; failed delivery is eligible for a refund.
Independent product. Not affiliated with, endorsed by or sponsored by Anthropic, OpenAI, Cursor or PayPal. No performance guarantees.