Notebook session ended
This session ended after eight hours.
Sign in again to continue from where you left off.
Notebook gate offline
The sign-in service is not responding right now.
Wait a minute and try again, or read what is already public from the hub.
Foundational walk through the so1 project. Read top to bottom if you're new; no prior context assumed.
More about this project on the So1 hub.
Use when: onboarding to the so1 control plane codebase; deciding where a new module belongs (UI, BFF, or shared); evaluating cross-repo dependency directions
Use when: adding a new BFF route exposed to the browser; debugging CORS or cookie failures on a fetch from rover; deciding whether a call belongs in a Server Component or via the proxy
Use when: debugging a redirect loop between /auth/login and a protected route; adding a new protected URL prefix; investigating why the session cookie is missing in production
Use when: adding a new BFF route that needs auth and logging; investigating which middleware mutated a response; tuning request-pipeline ordering
Use when: integrating a new external service into the BFF; writing a unit test that needs to stub upstream traffic; normalising heterogeneous error shapes into the so1 envelope
Use when: rendering job status in the UI; writing an assertion in a job-related test; deciding which states are terminal for log retention
Use when: designing a new job action and its inputs; writing redaction rules for sensitive inputs; querying jobs for SLA or audit reports
Use when: implementing the streaming log viewer in the UI; diagnosing reconnect/replay gaps in a long job; deciding how long to retain log lines after terminal state
Use when: writing a client retry loop for `POST /api/jobs`; investigating why a 'duplicate' job actually returned the same id; tuning the idempotency cache TTL
Use when: writing a new BFF route that returns an error; building UI error states that surface a support id; reviewing whether an upstream payload leaks secrets
Use when: choosing the correct code for a new failure path; writing client logic that branches on error semantics; auditing whether upstream errors are correctly classified
Use when: correlating a user-reported failure to BFF logs; wiring a new outbound call to a third-party API; validating that the OTEL trace id matches `requestId`
Use when: adding a new third-party integration that requires API tokens; auditing whether any secret can reach the browser; designing per-org token rotation
Use when: sizing instance counts for the BFF behind a load balancer; deciding whether an in-memory cache is acceptable in BFF code; configuring environment-specific secrets injection
Use when: writing the so1 Rover hero or one-liner; explaining what category Rover is in to a peer engineer
Use when: deciding whose objection wins when a screen has to choose; answering 'who is this dashboard actually for?'
Use when: writing copy for any Rover surface or social post; deciding whether a sentence sounds like Rover or like generic SaaS
Use when: explaining why Rover does not have a public sign-up yet; framing the path from dogfood to external operators for a stakeholder
Use when: pushing back on a 'add a hero metric block' suggestion; explaining what Rover is deliberately not