Back to browse
Specsight – Living product specs generated from your codebase

Specsight – Living product specs generated from your codebase

by aiola·Apr 12, 2026·2 points·0 comments

AI Analysis

●●SolidBig BrainSolve My Problem

Continuous spec sync detects changes from every release automatically.

Strengths
  • Context/Action/Outcome format structures specs for non-technical readers specifically.
  • Auto-detects changed files from commits and updates specs without manual triggers.
  • Product map visualizes feature connections with flow diagrams.
Weaknesses
  • AI code analysis for docs is crowded with Cursor, Cody, Mintlify.
  • Depends on Claude API costs at scale for large codebases.
Target Audience

Product managers and non-technical stakeholders

Similar To

Mintlify · Swimm · Sourcegraph Cody

Post Description

Heyy HN, I'm Ola

I'm an engineer myself, and everywhere I've worked there was a similar dynamic: someone from the non-technical side of the team (PM, CS, stakeholders) needs to understand what changed recently or how something works today. They rely on stale Confluence pages or ping engineers in Slack and wait. engineers get interrupted daily answering "how does this actually work". it's a mess that gets worse as the team grows

or when you join a new company and try to get a sense of the product. turns out the documentation doesn't exist or is months old. you ask 5 people and get 5 different answers. I've seen it and experienced it sooo many times

I decided to start exploring this space and built Specsight for those non-technical teams. it connects to your GitHub repos, analyses the codebase with Claude (Agent SDK), extracts business-level features of the product, and generates plain-language specifications in Context/Action/Outcome format for every feature. it also builds a visual graph of how features connect to each other, so you can see your product architecture without reading a line of code. On every merge or release, it re-analyses what changed and updates everything automatically

how it works under the hood: the AI does a full scan on initial setup, reads the entire codebase, then does merge syncs on each new release. a triage step filters out irrelevant merges so it only re-analyses what actually changed feature behaviour. It's capable of processing large codebases, though only on the paid plan (it's quite expensive to run heh)

tech stack for those curious: Next.js 16 on Vercel, Supabase (Postgres, Auth, Realtime), Claude Agent SDK for code analysis with a two-pass architecture (map the codebase then deep-dive each feature), Trigger.dev for background jobs, and a GitHub App for repo access and webhooks

ere's the live demo, a slice of Specsight's own spec generated from its own codebase: https://specsight.app/demo/project

I started just a few weeks ago, but I really want to spread the word about this and validate the idea. Hoping for your feedback and happy to answer any questions!

Similar Projects