docs: propose demand-driven session sidebar collections - #64
Conversation
Co-authored-by: jesse-merhi <79823012+jesse-merhi@users.noreply.github.com>
|
🦞👀 Pull request received. I will update this pull request when review starts. |
|
Codex review: needs real behavior proof before merge. Reviewed September 1, 2026, 1:57 AM ET / 05:57 UTC. ClawSweeper reviewWhat this changesThe PR adds a draft RFC proposing independently paginated, demand-driven session collections for the Control UI sidebar. Merge readiness⛔ Blocked until real behavior proof is added - 4 items remain Keep open for RFC acceptance: this is a structurally compliant draft that proposes a substantial Gateway and Control UI contract, with implementation intentionally deferred until maintainers accept the direction. Priority: P3 Review scores
Verification
How this fits togetherThe RFC repository records design decisions before implementation work begins. This proposal routes sidebar filters through Gateway-owned collection summaries and per-collection pages into a virtualized Control UI session list. flowchart LR
A[Sidebar filters] --> B[Gateway collection layer]
B --> C[Collection summaries]
B --> D[Per-collection pages]
C --> E[Sidebar manifest]
D --> F[Virtualized session sidebar]
E --> F
Decision needed
Why: The branch changes no runtime behavior, but it proposes a broad cross-component contract and explicitly leaves several policy and rollout choices unresolved. Before merge
Agent review detailsSecurityNone. Review metrics
Merge-risk optionsMaintainer options:
Technical reviewBest possible solution: Resolve the architecture questions through RFC review, then create a scoped implementation issue covering Gateway collection boundaries, Control UI migration slices, and browser proof before merging an accepted RFC. Do we have a high-confidence way to reproduce the issue? Not applicable: the branch adds a future-facing design document, not an executable behavior change. The referenced sidebar problem is outside this repository and was not independently reproduced here. Is this the best way to solve the issue? Unclear: the collection contract is coherent and explicitly preserves key boundaries, but the unresolved catalog, cursor, and rollout choices require maintainer acceptance before it is the settled implementation path. AGENTS.md: not found in the target repository. Codex review notes: model internal, reasoning high; reviewed against 967d9aac7472. LabelsLabel changes:
Label justifications:
EvidenceWhat I checked:
Likely related people:
Rank-up movesOptional improvements that raise the rating; they are not merge blockers.
Rating scale
Overall follows the weaker of proof and patch quality. Workflow
HistoryReview history (14 earlier review cycles; latest 8 shown)
|
Related: openclaw/openclaw#126701
Related: openclaw/openclaw#126708
Discussion: https://discord.com/channels/1456350064065904867/1541083365103370340
What Problem This Solves
The Control UI sidebar currently derives heterogeneous groups from a bounded
global session page. An expanded category can therefore look complete while
omitting sessions that fall beyond the global page, and loading another global
page can redistribute rows into several unrelated groups.
This RFC gives maintainers a concrete architecture to review before replacing
that behavior.
Why This Change Was Made
The proposal models the sidebar as an async tree: a lightweight root manifest
returns authoritative section summaries, while each expanded and visible
collection owns an independent cursor-backed child connection. A single
flattened virtual list controls rendered demand, and collapsed or distant groups
load no child rows.
The RFC also defines exact local counts, a canonical attention summary,
collection membership precedence, filter/cursor fencing, live invalidation,
bounded caching, failure states, migration steps, alternatives, and unresolved
thresholds.
The RFC remains
status: draft; its implementation issue stays blank untilmaintainer acceptance.
AI-assisted: yes. Jesse Merhi directed the product requirements and is credited
as an RFC author and commit co-author.
User Impact
This PR changes no shipped behavior. If accepted and implemented, operators
will get complete ordinary groups when they open them, accurate collapsed
counts and attention indicators, and no background loading for sidebar groups
they are not viewing.
Evidence
rfcs/0000-template.mdand includes all required frontmatterand top-level sections.
session-attention owners before defining the proposal.
Virtual, and the Relay cursor connection contract.
git diff --cached --checkpassed before commit.ownership rule, or unsupported new dependency.
authorize exporting the unpublished RFC before publication.