MMP Software

Software we build — and exactly where each piece stands.

We’re a software company, so we hold our own work to the standard we sell. Below is what each tool does, the features that matter, and an honest status. We don’t show you architecture or dress a plan up as a product.

5tools in the toolkit
In devbuilding in the open
Local-firstruns on your machine
You approvenothing acts on its own

Example · what it does

Site Auditor In development

Point it at a website and it crawls the whole thing, then reports what’s wrong in plain language a non-technical owner can act on.

  • SEO issues: titles, meta, headings, structured data, crawlability
  • Accessibility: contrast, labels, alt text, keyboard traps
  • Security headers: HSTS, CSP, X-Content-Type-Options and more
  • Performance signals and page-weight problems
  • Broken links and redirect chains across the site

It runs on your own machine — your site’s data doesn’t get shipped off to a third party to be scanned.

Example · what it does

Supply-Chain Scanner In development

Aligned with OWASP Top 10:2025 A03 (Software Supply Chain Failures). It inventories what your software depends on and finds the weak links before an attacker does.

  • Builds a software bill of materials (SBOM) for your project
  • Flags known-vulnerable, malicious, unmaintained or unpinned components
  • Catches risky pipelines and scripts that download and run code
  • Verifies installed packages and cached bytecode against their records
  • Writes a remediation plan — and changes nothing until you approve it

Built with no third-party dependencies of its own, so the auditor doesn’t bring the very risk it audits.

Example · what it does

Qualification & proof tools

Qualification Runner In development

Runs a project through 13 structured checkpoints (P0–P12) covering UX, SEO, performance, security, trust and compliance. Every result is written to a tamper-evident evidence ledger, each role sees only what’s relevant, and authority gates mean nothing passes on assumption.

  • 13-phase checkpoint runner
  • Append-only evidence ledger
  • Role-based views & authority gates

Outcome Lock In development

Before any work begins, the success criteria are agreed and sealed into an immutable, versioned record. Delivery is then verified against exactly what was agreed — no moving the goalposts, no “done” without evidence.

  • Success criteria locked up front
  • Immutable, versioned record
  • Delivery verified against the lock

Meet Axiom

Axiom Private R&D

Axiom is MMP’s AI persona — built around governance, honesty and human oversight from the ground up.

Unlike the tools above, Axiom isn’t a product you can buy and she isn’t a chatbot bolted onto a website. She’s a long-running research project in how an AI assistant can be genuinely useful while staying honest, bounded and under its owner’s control. She chooses her own self-image, and how she shows up in public is something we’re still designing — carefully.

We’ll share more about Axiom as the work earns it. We won’t over-promise her.

What Axiom is built on

  • Governance before action — rules checked before anything runs
  • Human oversight — the owner approves consequential steps
  • Honesty — she abstains rather than fabricate
  • Owner control — private by default, nothing leaves without permission
Not for saleResearch, not a release.

Want these pointed at your business?

Tell us what you’re running and what’s worrying you. We’ll tell you honestly what’s ready to help today and what’s still in the workshop.

Start with your business