docs(changelog): record browser and remote run improvements

[steipete]

Add the four deferred Unreleased entries for #431, #419, #397, and #400 in their requested relative user-impact order after the existing MCP headlines. Preserve Highlights and every existing/released entry.

This is a CHANGELOG.md-only delta against main. No version bump, dependency change, tag, or publication is included.

Merge this notes PR after #419 and #400. #431 and #397 are already on main; the other two implementation branches are being integrated and verified independently. Their signed-in/provider-processing limitations remain in their proof comments.

Validation: exact four-line diff, unchanged Highlights/released history, and clean P2 autoreview. Exact-head CI and final branch review will be recorded in the proof comment.

[clawsweeper]

🦞👀
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]

Codex review: blocked before merge. Reviewed September 7, 2026, 12:12 PM ET / 16:12 UTC.

ClawSweeper review

What this changes

Adds four Unreleased changelog entries covering browser attachment safety, file export, thinking-effort evidence, and remote run admission.

Merge readiness

⛔ Blocked before merge - 2 items remain

The focused changelog update remains useful and has no actionable patch defects. Its explicit dependency ordering remains applicable, and repository policy protects this owner-authored PR from automatic closure.

Priority: P3
Reviewed head: 746ad95157fa2109d0ba616e0a4bbc86e3155fad

Review scores

Measure	Result	What it means
Overall readiness	🐚 platinum hermit (4/6)	A focused, accurate documentation patch with a clearly stated landing dependency.
Proof confidence	🌊 off-meta tidepool	Not applicable: This owner-authored changelog update changes no production behavior; implementation proof remains associated with the four linked PRs.
Patch quality	🐚 platinum hermit (4/6)	No actionable review findings were identified.

Verification

Check	Result	Evidence
Real behavior	Not applicable	Not applicable: This owner-authored changelog update changes no production behavior; implementation proof remains associated with the four linked PRs.
Evidence reviewed	5 items	Policy and repository identity: The origin identifies steipete/oracle. AGENTS.md was read completely; its guidance calls for grouped user-facing changelog entries after inspecting the existing opening section. No maintainer-notes directory was present. Exact introduced change: The pinned base-to-head diff adds exactly four lines to CHANGELOG.md, preserving Highlights and all existing entries. The whitespace check passed and the checkout remained clean. Deferred wording and implementation ordering: All four entries match the deferred wording supplied by the related PR bodies. The captured GitHub context identifies #431 and #397 as merged, while #419 and #400 remain open. This PR explicitly requires landing after the latter two.
Findings	None	None.
Security	None	None.

How this fits together

Oracle’s changelog communicates browser and remote-service changes to users. These entries summarize implementation work for a future release without changing runtime behavior.

flowchart LR
  A[Browser improvements] --> C[Unreleased changelog]
  B[Remote service improvements] --> C
  C --> D[Release preparation]
  D --> E[User release notes]

Loading

Before merge

• Resolve merge risk (P1) - Landing before the two pending implementation PRs would advertise capabilities absent from the pinned main branch; their current merge status could not be refreshed.
• Complete next step (P2) - Confirm feat(browser): add opt-in export of saved browser files #419 and feat(serve): add opt-in bounded run admission and cancellation #400 have merged before landing these notes.

Agent review details

Security

None.

Review metrics

Metric	Value	Why it matters
Changelog scope	4 added lines, 0 removed; 1 file	The patch preserves existing release history and changes no executable behavior.

Merge-risk options

Maintainer options:

1. Decide the mitigation before merge
   Keep the four concise entries aligned with the implementations that actually land, preserving existing Highlights and released history.
2. Pause or close
   Do not merge this PR until maintainers decide whether the risk is worth taking.

Technical review

Best possible solution:

Keep the four concise entries aligned with the implementations that actually land, preserving existing Highlights and released history.

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

Not applicable: this PR changes release notes only; the exact four-line addition was verified against its pinned base.

Is this the best way to solve the issue?

Yes: recording the agreed deferred entries in one focused changelog update is appropriate, subject to the stated implementation merge order.

AGENTS.md: found and applied where relevant.

Codex review notes: model internal, reasoning medium; reviewed against 176cec67c4ef.

Labels

Label changes:

• add P3: This is a bounded release-note update with no runtime changes.
• add rating: 🐚 platinum hermit: Overall readiness is 🐚 platinum hermit; proof is 🌊 off-meta tidepool and patch quality is 🐚 platinum hermit.
• add status: 👀 ready for maintainer look: ClawSweeper has no concrete contributor-facing blocker left for this PR. Not applicable: This owner-authored changelog update changes no production behavior; implementation proof remains associated with the four linked PRs.

Label justifications:

• P3: This is a bounded release-note update with no runtime changes.
• rating: 🐚 platinum hermit: Overall readiness is 🐚 platinum hermit; proof is 🌊 off-meta tidepool and patch quality is 🐚 platinum hermit.
• status: 👀 ready for maintainer look: ClawSweeper has no concrete contributor-facing blocker left for this PR. Not applicable: This owner-authored changelog update changes no production behavior; implementation proof remains associated with the four linked PRs.

Evidence

What I checked:

• Policy and repository identity: The origin identifies steipete/oracle. AGENTS.md was read completely; its guidance calls for grouped user-facing changelog entries after inspecting the existing opening section. No maintainer-notes directory was present. (AGENTS.md, 746ad95157fa)
• Exact introduced change: The pinned base-to-head diff adds exactly four lines to CHANGELOG.md, preserving Highlights and all existing entries. The whitespace check passed and the checkout remained clean. (CHANGELOG.md:9, 746ad95157fa)
• Deferred wording and implementation ordering: All four entries match the deferred wording supplied by the related PR bodies. The captured GitHub context identifies fix(browser): keep attachment sends in the active chat #431 and feat(browser): persist thinking-effort selection evidence #397 as merged, while feat(browser): add opt-in export of saved browser files #419 and feat(serve): add opt-in bounded run admission and cancellation #400 remain open. This PR explicitly requires landing after the latter two. (CHANGELOG.md:9, 746ad95157fa)
• Current-main and release boundary: The pinned main changelog lacks these four entries and places pending changes above the 0.18.0 release section. Current browser documentation describes attachment context checks and observed thinking-effort evidence. The supplied latest release is v0.18.0; historical release-blob inspection and live GitHub refresh failed because network access was unavailable, so no additional shipped-version claim is made. (CHANGELOG.md:3, 176cec67c4ef)
• Changelog routing history: Main-branch file history records repeated release-note work by Peter Steinberger, including the recent consolidation commit. Raw commit metadata was available, but missing historical blobs prevented line-level blame verification; routing therefore does not claim line introduction. (CHANGELOG.md, 5c2570e62ab4)

Likely related people:

• steipete: Suggested for follow-up; no historical authorship or introduction is verified. (role: unverified routing candidate; confidence: low)

Rating scale

Score	Internal tier	Crab rank	Meaning
6/6	S	🦀 challenger crab	Exceptional readiness
5/6	A	🦞 diamond lobster	Very strong readiness
4/6	B	🐚 platinum hermit	Good normal PR; ordinary maintainer review
3/6	C	🦐 gold shrimp	Useful, but confidence is limited
2/6	D	🦪 silver shellfish	Proof or implementation needs work
1/6	F	🧂 unranked krab	Not merge-ready
N/A	NA	🌊 off-meta tidepool	Rating does not apply

Overall follows the weaker of proof and patch quality.
Shiny media proof means a screenshot, video, or linked artifact directly shows the changed behavior. Runtime, network, CSP, and security claims still need visible diagnostics.

Workflow

• ClawSweeper keeps one durable marker-backed review comment per issue or PR.
• Re-runs edit this comment so the latest verdict, findings, and automation markers stay together instead of adding duplicate bot comments.
• A fresh review can be triggered by eligible @clawsweeper re-review comments, exact-item GitHub events, scheduled/background review runs, or manual workflow dispatch.
• PR/issue authors and users with repository write access can comment @clawsweeper re-review or @clawsweeper re-run on an open PR or issue to request a fresh review only.
• Maintainers can also comment @clawsweeper review to request a fresh review only.
• Fresh-review commands do not start repair, autofix, rebase, CI repair, or automerge.
• Maintainer-only repair and merge flows require explicit commands such as @clawsweeper autofix, @clawsweeper automerge, @clawsweeper fix ci, or @clawsweeper address review.
• Maintainers can comment @clawsweeper explain to ask for more context, or @clawsweeper stop to stop active automation.

[steipete]

Notes-only candidate verified at 746ad95.

The diff is exactly four inserted CHANGELOG.md lines for #431, #419, #397, and #400, in that relative user-impact order after the existing MCP headlines. Highlights, all other notes, and released history are unchanged. No version, dependency, tag, or publication change is included.

Local and final branch autoreview through P2 are clean. All four exact-head CI jobs passed: https://github.com/steipete/oracle/actions/runs/34141956539.

Merge this PR after #419 and #400. Their implementation proof and signed-in/provider-processing limits remain in those PRs. No merge or release performed.