Skip to content

fix(ui): relabel agent owner attribution from "owned by" to "managed by" - #2133

Merged
tlongwell-block merged 1 commit into
mainfrom
eva/managed-by-label
Jul 19, 2026
Merged

tlongwell-block merged 1 commit into
mainfrom
eva/managed-by-label

Conversation

@tlongwell-block

Copy link
Copy Markdown
Collaborator

What

Relabels agent owner attribution across all user-visible surfaces from "owned by" to "managed by", per copy review in buzz-launch (2026-07-19).

Where

  • Desktop: message header owner treatment (MessageAgentOwner), mention autocomplete subtitles + titles, members sidebar add-member rows, new-message search result rows, profile popover owner line, profile panel "Managed by" field label (plus its ordering constant and testId user-profile-owned-by -> user-profile-managed-by; the testId had exactly one reference).
  • Mobile: compose bar mention suggestion subtitles (suggestions.dart), plus comments that quote the label.
  • Tests: e2e assertions in mentions.spec.ts, messaging.spec.ts, pubkey-display-screenshots.spec.ts updated to the new copy.

Not touched: backend/relay "owned by" strings about channels, tokens, and repos (ownership semantics, not the agent label), and code comments describing ownership behavior.

Testing

At 58b0a2f:

  • tsc --noEmit clean (desktop).
  • pnpm build then Playwright on the three touched specs: 87 passed, 1 failed — the failure (messaging.spec.ts "shows your avatar on your own message...") is untouched by this diff and passes solo; parallelism flake.
  • dart format + flutter analyze clean via pre-commit hooks (mobile).

"owned by" reads like a collar tag on something that talks back;
"managed by" is accurate and neutral. Applies to all user-visible
agent-owner copy: message header, mention autocomplete (desktop +
mobile), members sidebar, new-message search rows, profile popover,
and profile panel field label. Renames the internal label constant,
the user-profile-owned-by testId, and updates the e2e assertions to
match. Comments describing ownership semantics (not UI copy) are
unchanged.

Decided in buzz-launch: owner label copy review, 2026-07-19.

Co-authored-by: Tyler Longwell <tlongwell@block.xyz>
Signed-off-by: Tyler Longwell <tlongwell@block.xyz>
@tlongwell-block
tlongwell-block requested a review from a team as a code owner July 19, 2026 18:19
@tlongwell-block
tlongwell-block merged commit a1e977c into main Jul 19, 2026
25 checks passed
@tlongwell-block
tlongwell-block deleted the eva/managed-by-label branch July 19, 2026 18:40
BradGroux pushed a commit to BradGroux/buzz that referenced this pull request Aug 23, 2026
…by" (block#2133)

Signed-off-by: Tyler Longwell <tlongwell@block.xyz>
Co-authored-by: npub1qyvc0c5kl4gqv2fd97fsk46tu378sqgy35vc83rvgfwne90sel7s0ed67d <011987e296fd5006292d2f930b574be47c7801048d1983c46c425d3c95f0cffd@sprout-oss.stage.blox.sqprod.co>
Co-authored-by: Tyler Longwell <tlongwell@block.xyz>
tucktuck101 pushed a commit to tucktuck101/buzz that referenced this pull request Sep 17, 2026
…lock#2131)

The phase table claimed Phase 1 was "Not started". It shipped on 2026-09-08 as
PR block#2106, hardened by PR block#2133 on 2026-09-09. Both rows now state what is true
and name the merged PRs.

Phase 1b reads BUILT, NOT MERGED -- deliberately not "shipped". Every criterion
in its own review gate has been demonstrated with transcripts under
launchpad/plans/evidence/, but the branch has not been reviewed, gated or opened
as a pull request, and the row says so rather than letting a demonstration read
as a merge. The second-pass re-dispatch observable is specified and wired, but a
full two-pass run over a finished page belongs to Phase 2, the first phase that
produces one; that is stated rather than implied.

Phases 2-7 still read "Not started", correctly.

Five decisions recorded, under the honesty convention the rest of the document
uses -- what was decided, why, and what it does NOT solve. Open questions items
1-9 all closed on 2026-09-04, so these are a new section rather than new items:

  10. the re-dispatch observable, and its named independence limitation
  11. the roster-names redact-plus-flag shape, and its accepted contract cost
  12. column offsets as candidate identity, now demonstrated rather than argued
  13. what a missing headless CLI demonstrates -- consequences, not absence
  14. the verdict goes last in the response grammar, from measurement

Section 6.7's mechanism row said stdout was "captured as the verdict". That is
no longer precise: the whole response is matched against a two-line grammar by
equality, and completion is judged separately from content, so a non-zero exit,
a timeout or a truncated response each block whatever the text said.

Verified after editing, per this document's own house rules: every local section
cross-reference resolves to a real heading; the three references that do not
(2b, 2d, 4a) are to verify-claims/SKILL.md and each names that file immediately
before the reference; all three mermaid diagrams still present.

Signed-off-by: Serina Mcfall <serina.mcfall@gmail.com>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant