live system map

See what your agents actually built. Derivein crawls your repo into a deterministic map.

$ npx derivein crawl .
fig. 06 / run trail

When a run dies, you get an address.

System failures are mapped to the exact connection that failed, so the location is identified instantly.

1fetch_job()r1
2loadConfig()r2
3parse()r3 · died here
4transform()not reached
5saveResults()not reached
6notify()not reached
fig. 07 / honesty

Truth based on reality.

All visualized structure is computed deterministically, with zero AI opinion involved.

receipts hover or tap any edge
hole function · entryingest() functionnormalize() functionenrich() constSCHEMA functionpersist() key · event topiccomputed?
fig. 08 / your ai

The same map, served to your AI.

Your coding agent can also access derivein as a framework to code under, giving the agent a fuller, reality-based view of the code.

your agent asks
fig. 09 / connect

Two ways in.

path a · hosted

Connect a repo

A read-only GitHub App. Pick your repos, they are imported, and every push triggers a pull and a recrawl, so the map stays live with your work.

map updated
path b · local

Run the local agent

For code that never reaches GitHub: a zero-dependency agent crawls a local folder and syncs what it finds to your hosted map. Pairing is a one-time device token, and the sync heals itself after time offline.

$
synced · constant while online
the counterweight

Reviewers ask if the code is okay. Derivein asks if it is your idea.

Point derivein at a repo and watch the system assemble into a map you can hold in your head. The glance takes a minute, and it is the only way to see the decisions you never made.

$ npx derivein crawl .
now you can see