> For the complete documentation index, see [llms.txt](https://ai-os-and-trend-finder.gitbook.io/ai-os-and-trend-finder-docs/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://ai-os-and-trend-finder.gitbook.io/ai-os-and-trend-finder-docs/.spec_system/archive/phases/phase_33/session_01_capture_local_demo_runs.md). # Session 01: Capture Local Demo Runs **Session ID**: `phase33-session01-capture-local-demo-runs` **Status**: Not Started **Estimated Tasks**: \~12-25 **Estimated Duration**: 2-4 hours *** ## Objective Produce and review the real local Trend Finder and Dream Review outputs that will become the public demo source data. *** ## Scope ### In Scope (MVP) * Run the local Trend Finder capture path with reviewed public-source configuration and local credentials. * Use `bun run scheduler:trend-finder:run` or a full `bun run aggregate` run with `VITE_CLAUDE_OS_ENABLED_EXTENSIONS=trend-finder` or equivalent local extension enablement when refreshing Trend Finder. * Run the local Dream Review path against AI OS material suitable for public redaction. * Use `AI_OS_DREAM_ENABLED=true bun run scheduler:dream:run` for the private Dream run followed by `bun run aggregate`, or use `bun run scripts/dashboard-dream-run.ts` when the desired behavior is "run Dream and refresh LiveData after success". * Refresh `src/data/live-data.json` through the aggregate pipeline after the selected local runs complete. * Confirm `src/data/live-data.json` contains a populated `extensions.items["trend-finder"].data` branch. * Confirm the Trend Finder branch includes populated topics, evidence, sources, watchlist, Brief support, Engine Replay trace data, and advanced run artifacts when generated. * Confirm `src/data/live-data.json` contains a populated `dream` branch with meaningful prescriptions. * Confirm the Dream branch includes summaries, next actions, priorities, tags, and safe impact fields when available. * Record capture time, source commit, source set, source-set notes, provenance state, and high-level counts for both product areas. * Record Trend Finder counts: topics, evidence items, sources, watchlist rows, predictions, retros, and hidden gems. * Record Dream Review counts: prescriptions, categories, generated timestamp, and impact-field coverage. * Record whether Trend Finder analysis provenance was `ai-analyzed`, `deterministic-fallback`, `disabled`, or `mock`. * Record whether Trend Finder source provenance was `live`, `degraded`, `blocked`, or `fallback`. * Manually review selected Trend Finder and Dream content for demo quality and public-safety suitability. * Manually review top Trend Finder topics and evidence links for demo quality. * Manually review Dream prescriptions for usefulness, clarity, and public safety. ### Out of Scope * Committing public fixtures. * Changing exporter policy. * Deploying Cloudflare Pages. * Printing or documenting credential values. *** ## Prerequisites * [ ] Local runtime and source credentials are configured privately. * [ ] `src/data/live-data.json` can be refreshed locally. * [ ] Cloudflare Pages build remains unrelated to local capture commands. * [ ] Local-only scheduler, aggregate, and provider credentials or configuration are available when needed; none of these are part of the Cloudflare Pages build. * [ ] Local capture is run from this repository using local `.env.local` runtime and source credentials; values remain private and uncommitted. *** ## Deliverables 1. Refreshed local `src/data/live-data.json` containing the selected Trend Finder and Dream Review payloads. 2. Capture notes with counts, provenance, source commit, and manual review status. 3. Clear release-blocker note if `engineTrace` is missing or not sanitized. *** ## Success Criteria * [ ] Trend Finder has at least one strong public-evidence topic with a creator angle. * [ ] Hidden Gems, Sources, Watchlist, Brief, and Engine Replay have meaningful local content. * [ ] Local `engineTrace` is sanitized and includes non-empty timeline, section, source, evidence, runtime, artifact, or narration state. * [ ] The local Trend Finder branch includes non-empty `engineTrace` with state `sanitized`, non-empty decision timeline or section state, and non-empty source/evidence/artifact summaries. * [ ] The local Trend Finder branch includes populated `collectionHealth` and at least one of `runNarratives`, `demandClusters`, `industryEvents`, or `movementAnalyses` when the selected run generated those artifacts. * [ ] Dream Review has multiple useful prescriptions with clear summaries and next actions. * [ ] Reviewed source and analysis provenance are recorded accurately. * [ ] The source run is recent enough to be credible for the public demo. * [ ] Missing or non-sanitized `engineTrace` blocks release; recapture or fail the snapshot dry run instead of silently shipping `engineTrace` as missing. * [ ] The current local data candidate remains acceptable only after manual privacy review confirms populated Dream output, healthy Trend Finder collection state, and sanitized Engine Replay trace data. --- # Agent Instructions This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com. ## Querying This Documentation If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question. Perform an HTTP GET request on the current page URL with the `ask` query parameter, and the optional `goal` query parameter: ``` GET https://ai-os-and-trend-finder.gitbook.io/ai-os-and-trend-finder-docs/.spec_system/archive/phases/phase_33/session_01_capture_local_demo_runs.md?ask=&goal= ``` `ask` is the immediate question: it should be specific, self-contained, and written in natural language. `goal` is optional and describes the broader end goal you are ultimately trying to accomplish on behalf of the user. GitBook uses it to tailor the answer towards what is most useful for that goal. The response will contain a direct answer to the question and relevant excerpts and sources from the documentation. Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.