lda.chat/lda-wf
2
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

docs: plan scene 11-12 evidence and live activation

Browse Source
This commit is contained in:
lda
2026-07-11 21:32:50 +07:00 Verified
parent cbbb000a44
commit 6588c0edfe
3 changed files with 382 additions and 17 deletions
Download Patch File Download Diff File Expand all files Collapse all files
docs/current_roadmap.md
+12 -12
View File
@@ -177,22 +177,22 @@ Implementation order:
that never calls workflow authoring RPC operations during Scene 9
navigation. Implementation:
[`presentation agent authoring story`](historical/superpowers/plans/2026-07-11-presentation-agent-authoring-story.md).
26. Active: compress Scene 11 typed approval and Scene 12 resume/output/trace
evidence into a decision-led, continuation-led presentation, and restore
live prepared-run activation against the configured JSON-RPC target.
Implementation plan:
[`Scene 11-12 evidence and live activation`](superpowers/plans/2026-07-11-scene-11-12-live-evidence.md).
Presentation visual audit, July 10:
Presentation visual audit, July 11:
- Baseline-good: Scenes 3, 4, and 5 are good enough for now. Do not churn them
unless a later rehearsal exposes a concrete problem.
- Needs primary visual treatment: Scenes 1 and 2 still carry too much thesis
framing as text. They need a clear opening artifact: product title boundary,
problem shape, and "tool loop vs durable workflow" contrast.
- Needs cleanup, not reinvention: Scene 6 has the right interactive figure
foundation but still needs figure-specific polish where focus paths crowd or
hide labels.
- Needs visual strengthening: Scene 7 explains authoring but should show more
product evidence and less generic process language.
- Needs product-flow cohesion: Scenes 8-12 now have the right factual spine,
but the demo surfaces still need stronger hierarchy between chat, operation,
lifecycle, graph, interrupt, output, and evidence.
- Completed: Scenes 1, 2, 6, and 7 now have concrete focal artifacts, and
Scenes 8-10 share a prepared authoring/run spine.
- Active: Scenes 11-12 need evidence compression so approval, resume, output,
and trace remain factual without competing panels or repeated low-signal
values. The prepared live target must also remain live instead of being reset
to replay during route initialization.
- Completed: Scenes 13 and 14 now close with a bounded evaluation board,
contribution boundary/future-work map, and canonical defense-question index.