Skip to content

fix(crew): say a renamed crewmate keeps its short name - #26374

Merged
mbriones98 merged 2 commits into
mainfrom
fix/crew-short-name-hint
Oct 30, 2026
Merged

mbriones98 merged 2 commits into
mainfrom
fix/crew-short-name-hint

Conversation

@mbriones98

@mbriones98 mbriones98 commented Oct 30, 2026 •

Copy link
Copy Markdown
Contributor

Problem / Motivation

Goal: After a user names or renames a crewmate, the crewmate tells them that other crewmates still reach it by its short name.

The first crewmate's welcome asks "what would you like to call me?". The answer is saved as its display name only; its short name stays mate. When the user later asks another crewmate to hand work to the new name, select_crew answers unknown crew, and nothing along the way said the new name is a label.

Why it matters

The welcome invites a name, so most users rename their first crewmate. They find out the name does not work between crewmates only after memory and history have built up, when recreating the crewmate is no longer cheap.

Not a goal

  • Resolving a crew by its display name. The resolvers keep matching only the key, a case fold of the key and the member id.
  • Renaming a crewmate's short name.
  • Changing route_crew / select_crew beyond the unknown-crew answer.
  • Changing memory stores, member keys or routing.

What changed (motivation → approach → change)

The display name is a label by design, so this change makes the label visible as one rather than turning it into an identity.

  • The first-welcome intro (crewmate_prompts.json, filled by dashboard/mate_welcome.py) now tells the crewmate to say, right after it asks for a name, that other mates still reach it by its short name (mate).
  • The update_self tool description says a display name is a label only, and its rename reply names the short name from the gateway's answer and asks the crewmate to tell the user.
  • The crewmate prompt preview golden is refreshed to the current output. Besides this PR's sentence, that adds one Heartbeat sentence that was already in crewmate_prompts.json on main, where the golden had gone stale.

Backwards compatibility

Compatible: only prompt and tool-reply text changes; each adds a sentence.

Removes nothing: no field, path, key or capability is removed.

Tests

  • test/test_mate_welcome.py: the welcome says other mates still reach it by its short name; existing kickoff assertions fill the new {key} placeholder.
  • test/test_mate_update_self.py: the rename reply names the short name.

Proof: PROVEN test/test_mate_update_self.py::test_a_rename_says_other_mates_still_reach_it_by_its_short_name, test/test_mate_welcome.py::TestHiddenKickoff::test_the_welcome_says_other_mates_still_reach_it_by_its_short_name (head b6df261)

Manual verification

scripts/crewmate_prompt_preview.py --piece first_welcome renders the new sentence with mate; --fixture all-met --diff is clean after the refresh.

Related Issues

Closes #26370

Sibling sweep

Swept: git grep -n "display_name" -- src/kiro_crew/crewmate_prompts.json src/kiro_crew/mcp_mate.py
Sibling: src/kiro_crew/crewmate_prompts.json -- changed here

Pattern harvest

Not generalizable: a one-off missing sentence in a user-facing prompt, not a code pattern.

Checklist

  • At most two commits (one is the norm), with a Conventional Commits title
  • Existing tests pass and new tests added for new functionality
  • Self-review completed; code follows project style guidelines
  • Documentation updated (if applicable)
  • No secrets, credentials, or internal references in the diff

A crewmate's new name is saved as its display name only. Other
crewmates still reach it by its short name, and nothing told the user.

The first welcome now says, right after it asks for a name, that other
mates still reach it by its short name. The update_self description and
its rename reply say the same, naming the short name.

When select_crew is given a name that is exactly one crew's display
name, the unknown-crew answer carries a hint naming that crew's short
name. The hint only reports: the name is still refused, nothing is
bound or resolved, and a display name several crews share gets no hint.

Closes #26370
@mbriones98 mbriones98 self-assigned this Oct 30, 2026
@mbriones98
mbriones98 requested review from a team and KiroCrewCaptain as code owners October 30, 2026 19:36
@mbriones98
mbriones98 enabled auto-merge (squash) October 30, 2026 19:36
@github-actions github-actions Bot added the readiness: checking Automated validation is still running label Oct 30, 2026
@github-actions

github-actions Bot commented Oct 30, 2026 •

Copy link
Copy Markdown
Contributor

UX Review (Opus 5.5) — ✅ PASS

UX-level review of 886577c2b3cd6220da3c1b16a2ad6f6080ca6063 — updated in place on each push. A BLOCK verdict blocks PR readiness; PASS does not.

UX-Verdict: PASS

This PR only changes what the agent is told to say and one tool reply. No dashboard control or string changes, so no screenshot was needed.

Notes

  • welcome.intro: the new sentence explains how other mates reach this one (by its short name, mate) before the user has picked a name. The same message says "I'm Cove for now", so a first-time user meets two names at once, and "short name" is never explained. Better: drop this sentence from the welcome and keep only the reply that follows a rename in update_self, which is when the difference starts to matter.
  • update_self reply: "other mates still reach you by your short name" never tells the user what to do. Better: "To hand this mate work from another mate, call it {short_name}."

[UX-REVIEWED] 886577c

Description read

[DESCRIPTION-READ] 809227f16ca0f1838c9511b9abd05acef833901e06e22eeb241c9304d1635893

That sha256 names everything this verdict read: the PR description, plus the 3 evidence file(s) the workflow collected from it and handed this review. It covers that evidence because the media strip replaces every attachment URL with the same placeholder, so an edit swapping one attachment for another would leave the description's captured text identical while what this review judged changed. Recompute it with .github/scripts/pr-description-capture.sh over the same inputs; if it differs, the description or that evidence changed after this verdict was formed and any finding drawn from either is unproven.

@github-actions

github-actions Bot commented Oct 30, 2026 •

Copy link
Copy Markdown
Contributor

Design Review (Opus 5.5) — ✅ PASS

Design-level review of 886577c2b3cd6220da3c1b16a2ad6f6080ca6063 — updated in place on each push. A BLOCK verdict blocks PR readiness; PASS does not.

Design-Verdict: PASS

Nothing blocks: the change only adds prompt and tool-reply text that says the display name is a label, and select_crew resolution is left alone as the description promises.

Notes

  • The welcome hardcodes key=ASSISTANT_MEMBER_NAME instead of passing in the entry's real key. That holds only while welcome_kickoff stays Mate-only, which it is today (mate_welcome.py:335). If another crewmate ever uses this kickoff, it would tell users the wrong short name.

[DESIGN-REVIEWED] 886577c

Description read

[DESCRIPTION-READ] 05ef6b05ad9ac07bc314471963fd07fe04813cfe33261450935febd549d79b94

That sha256 names everything this verdict read: the PR description, plus the 1 evidence file(s) the workflow collected from it and handed this review. It covers that evidence because the media strip replaces every attachment URL with the same placeholder, so an edit swapping one attachment for another would leave the description's captured text identical while what this review judged changed. Recompute it with .github/scripts/pr-description-capture.sh over the same inputs; if it differs, the description or that evidence changed after this verdict was formed and any finding drawn from either is unproven.

@github-actions

github-actions Bot commented Oct 30, 2026 •

Copy link
Copy Markdown
Contributor

First Principles Review (Opus 5.5) — ✅ PASS

Premise-level review of 886577c2b3cd6220da3c1b16a2ad6f6080ca6063 — why this exists and whether the shipped surface is the smallest honest version. Updated in place on each push. A BLOCK verdict blocks PR readiness; PASS does not.

First-Principles-Verdict: PASS

Nothing blocks: every item that ships is one sentence of prompt or tool-reply text aimed at the reported surprise (#26370), and the resolvers are left alone, as ## Not a goal says.

Notes

  • Item 4 — rides along: "adds one Heartbeat sentence that was already in crewmate_prompts.json on main, where the golden had gone stale". The rider is declared and adds no surface of its own.

What this change ships

Inventory (4 items) — 3 justified

Intent: after a user names or renames a crewmate, the user learns that other crewmates still reach it by its short name. This is a FIX.

The supplied authentic.patch also carries main's own commits (the artifact download stash, the cloud boundary auto-upgrade, the Ultra Mode bolt and Board rename, the workflow tuple redaction). git diff a6f3cc5c HEAD shows the PR's real change is these 6 files, so those commits are not inventoried here.

  1. Mate's first welcome now says, right after it asks for a name, that other mates still reach it by its short name mate — justified
  2. The update_self tool description now says a display name is a label only and never changes the short name — justified
  3. After a rename, the reply names the short name (from the gateway's member field) and tells the crewmate to pass that on to the user — justified
  4. The prompt-preview golden is refreshed, which also picks up a stale Heartbeat sentence from main — rides along

[FIRST-PRINCIPLES-REVIEWED] 886577c

Description read

[DESCRIPTION-READ] 9d65ed40b3280d683c0054c66c3e929250a7583077ea51297a67a49dfd54db55

That sha256 names the PR description this verdict read. Recompute it with .github/scripts/pr-description-capture.sh; if it differs, the description changed after this verdict was formed and any finding drawn from the description is unproven.

@github-actions

github-actions Bot commented Oct 30, 2026 •

Copy link
Copy Markdown
Contributor

Opus 5.5 Review — ✅ no blocking findings

Reviewed 886577c2b3cd6220da3c1b16a2ad6f6080ca6063 — this comment is updated in place on each push.

Review details

No findings.

Notes:

  • The no-crew FIRST_WELCOME now tells the model to say "other mates still reach you by your short name" in the same message that calls this the user's first mate. This is prompt wording only and doesn't block.

[OPUS-REVIEWED] 886577c

Verdict parsed from the review's SHA-scoped output markers for commit 886577c2b3cd6220da3c1b16a2ad6f6080ca6063.

False positive or not applicable? A repository writer can comment:
/ai-review override fable 886577c2b3cd6220da3c1b16a2ad6f6080ca6063 false-positive: <reason, 20 to 500 characters>

@github-actions

github-actions Bot commented Oct 30, 2026 •

Copy link
Copy Markdown
Contributor

GPT 6.1 Review — ✅ human override accepted

Human judgment by @mbriones98 overrides the GPT 6.1 finding for 886577c2b3cd6220da3c1b16a2ad6f6080ca6063; the recorded reason is authoritative for this commit.

[GPT-OVERRIDE] 886577c

This comment is updated in place on each push.

The model was not re-run because an authorized human decision supersedes it.

False positive or not applicable? A repository writer can comment:
/ai-review override gpt 886577c2b3cd6220da3c1b16a2ad6f6080ca6063 false-positive: <reason, 20 to 500 characters>

The unknown-crew answer already lists every crew as its short name
beside its display name, so a separate hint field repeated it. The
welcome and rename text that tells the user about the short name stays.
@github-actions github-actions Bot added readiness: action required A blocking check or review needs attention and removed readiness: checking Automated validation is still running labels Oct 30, 2026
@mbriones98

Copy link
Copy Markdown
Contributor Author

/ai-review override gpt 886577c false-positive: the flagged decision entry is already on main from #26291; this PR changes only the six mate prompt, tool-reply and test files

@github-actions

Copy link
Copy Markdown
Contributor

Human judgment recorded

@mbriones98 recorded a false-positive release (class false-positive) for the gpt AI finding on 886577c2b3cd6220da3c1b16a2ad6f6080ca6063.

the flagged decision entry is already on main from #26291; this PR changes only the six mate prompt, tool-reply and test files

This decision applies only to this commit. A new push requires a new judgment.

@github-actions github-actions Bot added readiness: passed Eligible automated validation passed for the current revision and removed readiness: action required A blocking check or review needs attention labels Oct 30, 2026

@KiroCrewCaptain KiroCrewCaptain left a comment

Copy link
Copy Markdown
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Tier 1 auto-approve: fix (6 files). Criteria: no conflict, no requested changes, security path denylist clean, design-doc gate clean, security checklist all-NO, no reviewer-lane findings, readiness signal fresh for this head SHA, declared category matches the diff. Category: the welcome prompt and the update_self tool now tell a renamed crewmate that other mates still reach it by its short name, with prompt and tool tests updated.

@mbriones98
mbriones98 merged commit 857a62a into main Oct 30, 2026
97 of 106 checks passed
@mbriones98
mbriones98 deleted the fix/crew-short-name-hint branch October 30, 2026 23:06
@github-actions github-actions Bot removed the readiness: passed Eligible automated validation passed for the current revision label Oct 30, 2026
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.

A renamed crewmate is still reached only by its short name, and nothing says so

2 participants