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 sparki project. Read top to bottom if you're new; no prior context assumed.
More about this project on the Sparki hub.
Use when: onboarding to sparki-tools and deciding which repo a change belongs in; someone names auth-shield, deploy-loco, smithy or storm as if they were live; tracing a request from the browser to a build log
Use when: adding a capability to api-engine and deciding where it goes; someone cites subsystems/score or subsystems/loco as running code; debugging which package owns a failing API route
Use when: writing or reviewing a pipeline payload for POST /api/v1/pipelines; a step timeout or retry does not behave as the schema suggests; adding a field to PipelineStep in api-contracts
Use when: debugging a build that stays pending or fails with 'worker lost'; explaining how long a build may run and who stops it; tracing how a triggered build becomes logs and a final status
Use when: wiring the build log view to live output; debugging a dashboard that misses log lines or status changes; checking who may open a build's stream
Use when: writing a deployment status query; explaining why a deploy is `validating` vs `pending`; adding a new terminal status
Use when: choosing direct vs blue-green vs canary for a new service; investigating an unexpected auto-rollback; wiring a new platform adapter into deploy-loco
Use when: someone asks whether Sparki scans repositories today; re-enabling project detection behind the isolated worker; adding a language or framework to the detector set
Use when: adding a new metric type to the scoring engine; rendering a quality score on the dashboard; debugging a flaky-test detection result
Use when: deciding where a new engine signal should surface; the DSN sparki-engine scrape shows up == 0; someone cites observability-storm, ELK, Istio or Kubernetes as the sparki stack
Use when: writing the Sparki tagline or homepage hero; deciding whether a capability may be named in customer copy; explaining what category Sparki is in, today
Use when: writing audience copy for the landing page; briefing a designer on who the user is at the moment of contact
Use when: writing or reviewing any Sparki copy; deciding whether a tagline draft sounds like Sparki
Use when: explaining why Sparki publishes a design journal each sprint; framing the cross-channel cadence to a stakeholder
Use when: a design or copy review reaches for a comparison with another CI tool; answering 'how is this different from X' on a sales call; checking whether a landing-page block may mention a competitor
Use when: issuing, rotating or revoking a build-worker token; a worker gets 401, 403, 404 or 409 from /internal/worker/*; explaining what a stolen worker token can and cannot do
Use when: answering a prospect's question about how customer code is isolated; provisioning or verifying the build-worker VPS (gate G9); changing a sandbox limit, default image or the build network
Use when: support asks why Billing still shows Free after a successful checkout; changing a plan limit or adding a purchasable plan; preparing the live-mode switch (gate G-1-7)
Use when: a signed-in user gets 401 missing_session from the engine; wiring a new client (TUI, CLI, script) to api-engine; someone says BetterAuth or auth-shield still guards sparki
Use when: deciding where an env var, secret or DNS record for sparki lives; explaining why builds queue but never run; planning the G9 worker cutover
Use when: a sparki PR shows no GitHub Actions checks and you wonder if CI ran; running the CI suite for a sparki repo by hand; deciding where a new check belongs after the eu-1 retirement
Use when: preparing a prospect demo; a demo reset answers 409 or the demo login lands on an empty workspace; re-provisioning the demo login after an airlock change
Use when: someone says a so1 feature 'is part of sparki now'; planning work against the 2026-11-21 so1.io expiry; deciding whether a so1 repo may be archived
Use when: returning a new error from an api-engine handler; a client needs to branch on an engine error; comparing sparki's error shape with the retired so1 envelope
Use when: a customer reports an error and you need the matching log line; adding an outbound call from the engine; comparing sparki's correlation id with so1's requestId
Use when: adding an integration that needs a third-party secret; reviewing where a credential could leak to a browser or a build; comparing sparki's TOLL client and worker token with so1's adapters and secrets broker
Use when: the browser's API calls lose the session cookie or hit CORS errors; changing NEXT_PUBLIC_API_URL or adding a Next.js API route in web-app; comparing sparki's rewrite with so1's BFF route handler