w/01
AI-assisted code archaeology
We point Claude Code at your repo, ingest it systematically, and produce living documentation that reflects how the system actually behaves today — not how it was supposed to behave in 2019.
Service — Legacy Revival s/01
AI for the codebase nobody wants to touch.
We use AI to understand, document, and carefully modernize legacy systems — so the app that's been running your business for five years becomes maintainable, testable, and ready for the next decade. Without a rewrite.
01 / you'll recognise this
Symptoms.
- Your PHP 5.x / 7.x app still runs the business, but every deploy is a prayer.
- Your WordPress install has 40 custom plugins, no docs, and one contributor who left.
- Your microservice works in production but nobody remembers how it was supposed to work.
- Every ticket costs 3x what it should because understanding the code takes longer than fixing it.
- You've been quoted a rewrite. The quote is six figures. The risk is higher.
02 / the work
What we actually do.
w/02
Test coverage retrofit
We add behavioral tests for the paths that matter most, using AI to infer expected behavior from observed code. The tests protect you during the refactor and long after.
w/03
Safe, scoped refactors
We identify surgical wins — kill dead code, extract shared logic, enforce invariants — and ship them incrementally behind feature flags. Nothing big-bang.
w/04
Dependency and security audit
Outdated packages flagged, CVE exposure mapped, an honest upgrade path proposed. We say when the risk is worth it and when it isn't.
w/05
Modernization roadmap
What to touch first, what to leave alone, what's not worth the effort. Written for your team to execute — with or without us.
03 / the trade
What changes. What doesn't.
You keep
- Your business logic — nothing gets rewritten for aesthetics
- Your data and schemas — we work with what runs in production
- Your URLs, your SEO, your integrations
- Your team's mental model — we document what they already know
You get
- Living documentation, updated alongside any change we ship
- A test suite that catches real regressions, not just the easy ones
- A maintainability baseline — so you can measure improvement
- An actionable roadmap your team can execute without us
04 / engagement
Where this fits.
Legacy Revival is the implementation side of the engagement ladder. The ladder itself is the same for every service we run:
- step 01 Free Snapshot
We review your site or repo and send a 1-page report with 3–5 AI integration opportunities specific to Legacy Revival. 2 business days. $0.
- step 02 Full Audit — $1,200 flat
One week deep-dive. 10–15 page prioritized roadmap with estimates, risks, and an implementation plan. Credited against the project if we work together.
- step 03 Implementation — $4K–$25K
Scoped work, fixed price, clear deliverable. Most Legacy Revival projects land between $6K and $15K.
05 / stack
What we reach for.
- Languages
- PHP · Python · Ruby · Node.js · Go · Java
- Frameworks
- Laravel · Symfony · WordPress · Django · Rails · Express
- AI tooling
- Claude Code · custom MCP servers · AST-aware refactor agents
- Testing
- Playwright · Pest · PHPUnit · Jest · Vitest
- Runtimes
- on-prem · VPS · Docker · Kubernetes
These are examples, not rails. We pick tools per engagement based on what already lives in your stack — we don't force a preferred tech on you.
06 / questions
Questions we hear a lot.
q/01 Can you work with a codebase you've never seen?
That's the point. Legacy Revival starts from the assumption that we know nothing and the codebase has to teach us. The AI-assisted archaeology is how we learn fast without making assumptions.
q/02 How do you avoid breaking production while refactoring?
Behavioral tests first, refactors second. Every scoped refactor ships behind a feature flag, gets compared against pre-change behavior, and rolls out gradually. If something breaks, it breaks in staging, not at 2am.
q/03 What if my team thinks AI-generated documentation is going to be wrong?
They're right to worry. We don't publish AI output unreviewed — every doc gets a human pass (by us, initially, and by your team as the engagement matures). The AI is scaffolding, not the final product.
q/04 Do you need access to production?
No. Access to the repo and a reproducible local or staging environment is enough. For the Snapshot, a public URL or read-only repo access is sufficient.
q/05 We've been quoted a rewrite. Why would we do this instead?
A rewrite replaces risk with bigger risk, on a longer timeline, for more money. Legacy Revival makes the existing system work better for a fraction of the cost. If after the Audit we conclude that a rewrite really is the right answer, we'll tell you — and we won't be the ones to do it.
q/06 What if the code is too bad to save?
Some codebases genuinely do need to be rewritten. The Audit is where that call gets made, with numbers. Roughly one in five engagements we're honest and say 'this one should be rewritten' — and we help scope that, but we won't sell you a rewrite we don't believe in.
Start with a Snapshot.
Two business days. No strings. A 1-page report with 3–5 AI integration opportunities specific to your situation.