[ Idea ]Found 12 Nov · 3 Ask HN

Handoff

Independent health reports that prove vibe-coded apps are sound.

Should you build this?Don't build

VibeDoctor and quality·vibes already scan vibe-coded repos with shareable reports, so Handoff is a copy.

7 products · 9 sources13 Nov

The problem

Code built with AI assistants decays over time, and developers have no credible way to prove to clients that it is sound.

Non-technical clients cannot judge code quality, and the current options are one-off human audits costing $199 to $3,500 or enterprise tools built for engineering teams. Freelancers and small agencies lose deals and trust as a result.

Seen in 3 Ask HN

The idea

Handoff scans the project on the developer's machine and produces a plain-English, tamper-evident health report that the client can read and trust.

1Run the Handoff command in your project folder.
2Only metrics upload. Your source code never leaves your machine.
3Share a link with your client showing trends and a plain verdict.

Price: First report free. $29 per milestone report (about ₹2,400). $19/month per tracked repo for automatic snapshots and a shareable client link. Optional $99 human spot-check upgrade, only after demand is proven.

[ The six checks ]Failed at check 3 of 6

1Not built in alreadyPassed

2An AI agent can't do itPassed

3Nothing similar existsFailed here

4People pay for thisNot checked

5Strong scoresNot checked

6Real sourcesNot checked

1 · Not built in alreadyPassed

2 · An AI agent can't do itPassed

3 · Nothing similar exists

Why it failed

Check 3 · Already exists

VibeDoctor and quality·vibes are live automated tools that already score vibe-coded repos for code health, duplication and complexity. Both also offer shareable reports and recurring or on-push scans. Handoff's remaining differences (hash-stamped history, cohort percentile, client-facing framing) are thin and easy for them to add.

Who paysThe freelancer or agency pays $29 per milestone report, or $19/month per tracked repo, because the report helps them close and keep clients. Later, non-technical founders and micro-SaaS buyers who commission the work will also pay.

Who uses itFreelancers, small agencies and student dev shops building client apps with Claude Code, Codex or Kiro. Solo builders who want to prove their repo is not rotting.

First 10 users1) On X (~750 builder followers), post 'health curves' of 5 well-known open-source vibe-coded repos. The tool is free for public repos, and this shows the product working. 2) Offer free reports to freelancers in Indian builder communities, Superteam India's freelance members and the LinkedIn network, in exchange for sending the report to a real client. 3) Offer hackathon organisers and DevRel contacts a free 'build health scorecard' for submissions, as distribution. 4) DM the 10 freelancers who send reports to clients and ask for $29 on the next milestone.

Why not just AIThe metrics can be reproduced by an agent, but the trust cannot. A client will not accept a freelancer's own agent grading its own work. Handoff gives a third-party, hash-stamped snapshot per commit SHA, so the history cannot be quietly rewritten. It also gives a percentile against a pooled, anonymised cohort of other agent-built repos ('your duplication is worse than 80% of similar repos'), and an agent working on one repo cannot produce that. The buyer is also often non-technical and has no agent workflow for judging code.

What already exists

7 products checked

ProductLocal-first scanningSix-signal health scoreAutomatic milestone snapshotsShareable client report linkOverlap with idea

This ideaWhat Handoff plans to do

4 / 4

quality·vibes (ogbuilds.ai)Scores code cleanliness and security with weekly monitoring, a shareable report, a README badge and pay-per-report pricing. · the reason it failed

[ What would make it pass ]3 steps

01
Check VibeDoctor and quality·vibes pricing, features, and report pages first.
02
Find a moat they cannot copy quickly, such as a real anonymised cohort.
03
Get ten freelancers to pay for a report before building more.

[ Inspired by ]What it borrows, and what's different

CodeScene (Code Health + trend)
Behavioural code-health score with a historic trend, priced per author for dev teams. Handoff is built for freelancer-to-client handover, not team management. It gives a plain-English verdict for non-technical readers, a hash-stamped history, and flat per-repo pricing instead of per-author seats.
GitClear AI code quality research
Large-scale duplication, churn and refactoring signals across AI-assisted commits. GitClear sells enterprise ROI dashboards. Handoff turns similar signals into a per-repo percentile for small builders, using a cohort that fills from users' anonymised metrics.
Valletta Vibe Coding Audit
A fixed-price human audit of AI-built apps with a prioritised fix list. Handoff is automated, so it costs about $29 instead of $199 or more. It can be re-run at every milestone, so decay shows up as a trend rather than a one-off.

[ Sources ]9 cited

A current HN thread has a dev who reviewed and architected everything by hand now struggling to keep quality with Claude. Replies suggest ad-hoc fixes such as asking an AI to audit now and then, which is no reliable check.news.ycombinator.com ↗
GitClear's June 2026 report finds that, across hundreds of millions of changes, block duplication rose 81% since 2023 and refactoring fell sharply. This is the aggregate signal Handoff would track per repo.gitclear.com ↗
There is a live flood of reports that Opus 5.5 got worse, so developers cannot assume the quality of agent output is stable and need their own measurement.news.ycombinator.com ↗
A related Ask HN asks how to evaluate agent skills and finds people mostly read the SKILL.md and try things out. The same lack of systematic checking applies to agent output.news.ycombinator.com ↗
A fixed-fee human audit of AI-built apps is sold from $199, with a stated aim of giving investor-ready, due-diligence-proof code. This shows people pay for a manual version of this report.vallettasoftware.com ↗
Another vendor sells a $149 vibe-coded app audit, with the fee credited against fix work.codechoicez.com ↗

Why it failed

Check 3 · Already exists

What already exists

7 products checked

ProductLocal-first scanningSix-signal health scoreAutomatic milestone snapshotsShareable client report linkOverlap with idea

This ideaWhat Handoff plans to do

4 / 4

quality·vibes (ogbuilds.ai)Scores code cleanliness and security with weekly monitoring, a shareable report, a README badge and pay-per-report pricing. · the reason it failed

Code built with AI assistants decays over time, and developers have no credible way to prove to clients that it is sound.

Handoff scans the project on the developer's machine and produces a plain-English, tamper-evident health report that the client can read and trust.

Why it failed

What already exists

Code built with AI assistants decays over time, and developers have no credible way to prove to clients that it is sound.

Handoff scans the project on the developer's machine and produces a plain-English, tamper-evident health report that the client can read and trust.

Why it failed

What already exists

How it was checked