Skip to content

Gateway capability tracking #1829

Open
Open

Description

@roboclaw-bot

Permanent overview for upstream Gateway capability investigations in the Windows companion. Keep this issue open; do not assign it to Copilot.

The monitor will update only the marked section below. Human notes outside it and all comments are preserved. This issue holds the current investigation, durable task state, disposition, and links to any resulting draft PRs. Legacy investigation #1845 is closed as a preserved evidence archive, not an acknowledged finding. Repeated unchanged observations should not post comments or create replacement trackers.

Implementation: #1692

Monitoring is enabled. The single-tracker and separate Copilot authentication implementation is deployed by PR #1847. The monitor records a confirmed task ID below when a live handoff succeeds.

Capability review status

Investigation only. Main/schema evidence does not establish released-source compatibility or complete Windows support.

question-answer: deferred-active-task.
Candidate fingerprint: 6125d6f0a5de9832199dff93bc39af4277bc929e170881109a84eb3635bd62ea.
No confirmed task ID. No automatic retry after a reserved attempt.

Investigation evidence.

Monitor needs attention: failed run. No new compatibility conclusion.

Activity

  1. added
    P3Low-risk cleanup, docs, polish, ergonomics, or speculative feature.
    on Oct 22, 2026
  2. clawsweeper commented on Oct 22, 2026

    @clawsweeper

    Codex review: this still needs some work.

    Summary
    Keep this issue open: deployed automation requires this permanent tracker, and the latest run at current main still reports collection failures.

    Reproducibility: This is a permanent operational tracker rather than a single bug report; hosted results verify current-main collection failures, and inspected source and tests establish its required lifetime without local execution.

    Root-cause cluster
    Relationship: canonical
    Canonical: #1829
    Summary: This is the designated permanent tracker for capability investigations; the archived candidate explicitly migrated here.

    Members:

    Proposal only: this assessment does not dispatch repair, suppress jobs, mutate sibling items, close, or merge anything.

    Next step
    Diagnose the protocol collection exception, establish a pinned failing regression for the moved welcome hint, and refresh the deferred question assessment while preserving this tracker.

    Review details

    Best possible solution:

    Preserve this tracker with reliable collection and current, evidence-backed capability dispositions.

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

    This is a permanent operational tracker rather than a single bug report; hosted results verify current-main collection failures, and inspected source and tests establish its required lifetime without local execution.

    Is this the best way to solve the issue?

    The deployed single-tracker contract remains appropriate; establish failing regressions from pinned collection inputs before repairing readers, and reassess the old question disposition against current implementation.

    AGENTS.md: found and applied where relevant.

    Remaining risk / open question:

    • The protocol collection exception remains unidentified: annotations report an unsuccessful exit, while detailed log retrieval was blocked by the artifact-host network allowlist.
    • The deferred candidate has no confirmed task ID, and current source does not establish released-Gateway compatibility.

    Codex review notes: model internal, reasoning medium; reviewed against 91070a21147a.

    Label changes

    Label changes:

    No label changes.

    Label justifications:

    • P3: This is an internal monitoring tracker; its collection failures do not establish a blocked end-user workflow.
    Evidence reviewed

    What I checked:

    How this review workflow works
    • 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; those execution lanes are retired.
    • Maintainers can comment @clawsweeper explain to ask for more context, or @clawsweeper stop to stop active automation.

    Reviewed November 22, 2026, 12:52 AM ET / 05:52 UTC.

  3. github-actions commented on Oct 24, 2026

    @github-actions
    Contributor

    Welcome-copy monitoring baseline

    The first enabled observation records a baseline, not a claim that Windows wording already matches upstream.

    Upstream source

    This is upstream-main presentation copy, not released Gateway compatibility evidence. Keep welcome text bundled; do not add a Gateway RPC. This notice does not start an agent or change the app.

    Monitor run

  4. github-actions commented on Oct 29, 2026

    @github-actions
    Contributor

    Copilot investigation result

    Completed task. Final answer retrieved by the monitor, not independently verified runtime proof.

    Disposition: existing-work. No code changes or duplicate PR were made.

    The immutable upstream main commit (021bdf1d…) and npm protocol artifact were independently checked. The npm tarball’s SHA-512 integrity matched; its canonical schema hash is 90d43d4f…. Released Gateway source compatibility remains unverified.

    Current Windows production lacks question RPCs and lifecycle-event dispatch. Open PR #1757 contains the dedicated question flow and fixture-backed behavioral tests; #1835 explicitly leaves question integration to it. Those tests and real-Gateway behavior are not fully validated.

    Important unresolved check: the protocol requires operator.questions, but current client scope requests and #1757’s diff do not add it. The interactive fixture advertises that grant, so it does not prove real scope negotiation. This is not a claim of supported or released-source compatibility.

    I could not post the requested tracker comment: gh requires an unavailable GH_TOKEN, and the browser tool’s transport was closed. The exact disposition marker to include is:

    <!-- upstream-gateway-disposition:v1:9a9c182b64d1d6e0c376c16622c4862323b6bf8a5d47853673cc71428dd3ed8f:existing-work -->

    No repository validation was run because no files were changed.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Metadata

Metadata

Assignees

No one assigned

    Labels

    P3Low-risk cleanup, docs, polish, ergonomics, or speculative feature.issue-rating: 🌊 off-meta tidepoolIssue quality rating does not apply to this item.

    Type

    No type

    Fields

    Priority

    None yet

    Projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions