Skip to content

fix(console): say the dedicated reason once and drop the repositories promise on create agent - #3573

Merged
freeqaz merged 2 commits into
mainfrom
fix-d700-create-agent-copy
Oct 25, 2026
Merged

freeqaz merged 2 commits into
mainfrom
fix-d700-create-agent-copy

Conversation

@freeqaz

@freeqaz freeqaz commented Oct 25, 2026

Copy link
Copy Markdown
Collaborator

Problem

Create Agent copy and spacing issues from dogfood D700 (board finding 2404):

  • The start page intro said "Choose a model, connect repositories, and give your Agent a place to work." on an Installation without repository access.
  • With no Sandbox Driver, the Dedicated OpenClaw reason appeared twice: in the Harness hint ("This installation selects no Sandbox Driver, so choose Codex for Dedicated execution.") and in the Execution mode hint under Runtime details.
  • At 390 px the two start cards had different inner padding. The intro card kept 24 px; the Preset card used the card's phone padding of 16 px.

Change

  • The intro now reads "Choose a model and give your Agent a place to work." The form shows repository access only where the Installation sets it up.
  • The Harness hint says only what this Agent will run and that Codex gives Dedicated execution: "This Agent will run OpenClaw with Embedded execution. Choose Codex for Dedicated execution." The reason stays once, under Runtime details, as finding 1904 intended.
  • .launch-intro no longer overrides the card padding, so both start cards use it at every width.
  • The Storybook scenario createOpenclawWithoutNativeWorkers has a step for the single reason.

This does not touch the repository selection or notConfigured paths (#3568/#3569, fix-postland-1025b).

Verification

  • console-page-header.test.ts "Create Agent leads with ← Agents…" checks the intro text and that both cards have the same padding at 1440 and 390. console-create-agent-provisioning-400.test.mjs (two tests) checks the new Harness hint text. All three fail on main and pass here: 20/20 at 2 CPUs on a remote test VM.
  • checks-browser and checks-browser-2 green on the branch on the remote test VM.
  • Before and after (simulated console fixture; a Kubernetes Compute fixture without a Sandbox Driver):

Before, start 390:
d700-before-start-390
After, start 390:
d700-after-start-390
Before, OpenClaw 1440:
d700-before-openclaw-1440
After, OpenClaw 1440:
d700-after-openclaw-1440

Video (after: the start page, then OpenClaw with Runtime details): d700-after.webm

Risks

Copy and one CSS line. The Harness hint still names Codex only while the harness can change.

@freeqaz
freeqaz requested a review from a team as a code owner October 25, 2026 03:19
@clawsweeper

clawsweeper Bot commented Oct 25, 2026 •

Copy link
Copy Markdown
Contributor

🦞👀
ClawSweeper picked this up.

Pull request received. I will update this pull request when review starts.

ClawSweeper review complete

ClawSweeper finished reviewing this revision. The review result is being finalized.

View the workflow run.

@clawsweeper clawsweeper Bot added clawsweeper:intake:bug Preliminary submission classification; full source review pending. clawsweeper:intake:needs-environment Preliminary submission classification; full source review pending. clawsweeper:intake:needs-repro Preliminary submission classification; full source review pending. P3 Low-risk cleanup, docs, polish, ergonomics, or speculative feature. proof: sufficient Contributor real behavior proof is sufficient. proof: 📸 screenshot Contributor real behavior proof includes screenshot evidence. rating: 🐚 platinum hermit Good normal PR readiness with ordinary maintainer review expected. status: 👀 ready for maintainer look ClawSweeper has no concrete contributor-facing blocker left for this PR. and removed clawsweeper:intake:bug Preliminary submission classification; full source review pending. clawsweeper:intake:needs-environment Preliminary submission classification; full source review pending. clawsweeper:intake:needs-repro Preliminary submission classification; full source review pending. labels Oct 25, 2026
@clawsweeper

clawsweeper Bot commented Oct 25, 2026

Copy link
Copy Markdown
Contributor

Codex review: needs maintainer review before merge.

What this changes

This PR removes an unconditional repository promise, shortens the Harness hint, and aligns Create Agent card padding on phones.

Example: Create Agent at 390 px on an Installation without repository access

  • Before: The introduction promises repository connections and has 24 px padding while the Preset card has 16 px.
  • After: The introduction promises model and workspace setup, and both cards use 16 px padding.

Review scores

Measure Result What it means
Overall readiness 🐚 platinum hermit (4/6) A narrow, correct presentation repair supported by directly inspected before/after Console screenshots and focused browser coverage.
Proof confidence 🐚 platinum hermit (4/6) ✨ media proof bonus Sufficient (screenshot): Inspected before/after screenshots exercise the rendered Console at 390 px for the start cards and 1440 px for OpenClaw with Runtime details expanded, showing the changed text and spacing. These are simulated Console fixtures, sufficient for this presentation-only change; they establish no backend or runtime behavior. The linked recording could not be inspected because it was omitted from prepared media and subsequent access was blocked by review infrastructure.
Patch quality 🦞 diamond lobster (5/6) No actionable review findings were identified.

Product

Kind: Bug fix · Worth it: Yes · Fix scope: Complete
User problem: Create Agent promises repository access that may be unavailable, repeats the Dedicated OpenClaw explanation, and gives its phone start cards inconsistent padding.
Reason: This is a small, maintainer-authored correction to observed Console presentation problems, with the existing capability explanation preserved.

Merge readiness

✅ Ready for maintainer review

Keep this PR open: it is a focused, maintainer-authored Console correction, with no actionable introduced defect found.

Priority: P3
Reviewed head: 38e29c041dc9c20f46d36b54800e762bb2ea9a6b

Before merge

None.

Findings

None.

Agent review details

How this fits together

The Console renders Installation capabilities and current form choices into Agent setup guidance; this change affects displayed text and spacing.

flowchart LR
  A[Installation capabilities] --> B[Create Agent form]
  C[Provider and Harness choices] --> B
  B --> D[Harness outcome hint]
  B --> E[Runtime details explanation]
  F[Shared card styles] --> G[Create Agent start cards]
Loading

Technical review

Best possible solution:

Keep the capability-specific explanation under Runtime details and let both start cards inherit their shared responsive padding.

Do we have a high-confidence way to reproduce the issue?

Current-main source and supplied before screenshots establish the copy and spacing problems; no target code or tests were run during this read-only review.

Is this the best way to solve the issue?

The patch uses the existing text owners and shared card styles, preserves capability checks and locked-Harness guidance, and adds no unnecessary production machinery.

AGENTS.md: found and applied where relevant.

Codex review notes: model internal, reasoning medium; reviewed against 177bd674bb1a.

Provenance checked

  • Start-card introduction and launch styling changes intended behavior with a stated reason (354e94b: The launch simplification introduced progressive disclosure and the current start-card presentation; the subsequent token refactor preserved its padding override.)
  • Harness outcome versus installation conditions keeps the original intent (6b0a79c: The Harness hint states what this Agent will run, while Runtime details retains Installation conditions.)
  • Precise no-Sandbox-Driver explanation changes intended behavior with a stated reason (2b1f510: Replace a hedged explanation with the known reason Dedicated OpenClaw is unavailable; this PR explicitly removes its repetition while retaining that reason under Runtime details.)

Testing

Proof path: shipped entry point.

Security

None.

Evidence

What I checked:

Labels

Label changes:

  • add P3: The change corrects misleading and repeated guidance plus responsive spacing without changing the Agent creation workflow.
  • add proof: sufficient: Contributor real behavior proof is sufficient.
  • add proof: 📸 screenshot: Contributor real behavior proof includes screenshot evidence.
  • add rating: 🐚 platinum hermit: Overall readiness is 🐚 platinum hermit; proof is 🐚 platinum hermit and patch quality is 🦞 diamond lobster.
  • add status: 👀 ready for maintainer look: ClawSweeper has no concrete contributor-facing blocker left for this PR.

Label justifications:

  • P3: The change corrects misleading and repeated guidance plus responsive spacing without changing the Agent creation workflow.
  • rating: 🐚 platinum hermit: Overall readiness is 🐚 platinum hermit; proof is 🐚 platinum hermit and patch quality is 🦞 diamond lobster.
  • status: 👀 ready for maintainer look: ClawSweeper has no concrete contributor-facing blocker left for this PR.
  • proof: sufficient: Contributor real behavior proof is sufficient.
  • proof: 📸 screenshot: Contributor real behavior proof includes screenshot evidence.

Rating scale

6/6 🦀 challenger crab · 5/6 🦞 diamond lobster · 4/6 🐚 platinum hermit · 3/6 🦐 gold shrimp · 2/6 🦪 silver shellfish · 1/6 🧂 unranked krab. Overall follows the weaker of proof and patch quality; ✨ marks media proof (a screenshot, video, or linked artifact) that directly shows the changed behavior.

Workflow

ClawSweeper edits this one comment on every review. Comment @clawsweeper re-review for a fresh review only. Repair and merge execution lanes are retired.

Reviewed October 24, 2026, 11:32 PM ET / October 25, 2026, 03:32 UTC.

@freeqaz
freeqaz merged commit 44eebaa into main Oct 25, 2026
84 checks passed
@freeqaz
freeqaz deleted the fix-d700-create-agent-copy branch October 25, 2026 03:35
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

P3 Low-risk cleanup, docs, polish, ergonomics, or speculative feature. proof: 📸 screenshot Contributor real behavior proof includes screenshot evidence. proof: sufficient Contributor real behavior proof is sufficient. rating: 🐚 platinum hermit Good normal PR readiness with ordinary maintainer review expected. status: 👀 ready for maintainer look ClawSweeper has no concrete contributor-facing blocker left for this PR.

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant