FAQ
No marketing fluff or scripted support responses. If you still have questions after reading this, email matt@codewheel.ai for a direct response.
Matt Owens — former Tesla Staff Engineer with 15 years of production engineering experience. CodeWheel builds open-source developer tools (PromptWheel, MCP packages) and publishes technical content on AI architecture and security.
We publish technical content and case studies so you can evaluate the work directly. Our founder's 15-year track record is verified on LinkedIn, and we're happy to walk through past architectures on a call.
Open-source developer tools (PromptWheel, MCP error codes, MCP PHP toolkit, MCP Tools for Drupal) and technical content covering AI architecture, agent security, prompt injection testing, and production patterns.
Next.js, Astro, Supabase, Postgres/pgvector, Stripe, Clerk, Vercel, AWS, OpenAI/Anthropic, and a long list of supporting security tools (OWASP ZAP, Burp, Nuclei, Kali, etc.).
All our packages are on GitHub. Stars, issues, and pull requests are welcome. Check out PromptWheel, MCP error codes, or MCP Tools for Drupal.
Tools are permissioned, inputs are schema-validated, and sensitive actions require confirmations or human-in-the-loop checks. We also run offensive tests against tool bindings before they ever hit production.
Row-level security, per-tenant namespaces for embeddings, and strict filters at the retrieval layer. Every RAG and agent call is scoped before it ever reaches the model.
Prompt-injection suites, RAG retrieval boundary tests, and Playwright E2E flows run in CI. Offensive tests happen in a sandbox before go-live, and we re-test after fixes are applied.
Check out our open-source tools, read the blog, or drop us a message.
