Self-driving 101
Self-driving is PostHog using your product data as context to make your product better. Here's how it works: Scouts watch your data on a schedule, file reports when something needs your attention, and an agent opens a pull request that fixes it. You review and merge everything, so nothing ships without you.
The loop
The whole system is Fig. 1. A signal is something in your product data that's worth knowing, with its evidence attached. Related signals group into a report for you to review. An agent writes the fix as a pull request, and PostHog measures whether it worked. The outcome becomes new signals, so your product keeps improving itself over time.
the outcome becomes new signals
High-confidence reports arrive with the pull request already open – you just review and merge. Lower-confidence ones wait in your inbox for you to decide first.
When a scout writes
A scout run ends one of two ways: it writes a report, or it writes nothing.
A scout writes a report when it spots a change that needs your attention – your AI bill jumped, or your core activation metric dipped. It stays quiet when everything looks healthy, including when a change is just growth.
Quiet is the common case. A scout that reports every day is one you'll mute before you get real value out of it.
How to read the use cases
Each use case in this pocket guide is a real custom scout you can add to your own product. Every guide covers:
- the scout's job, and the anatomy of the inbox report it files
- the pull request that report becomes
- the scout itself – the actual file an agent follows
- a one-click way to add it to your own product