Skip to content

feat(dashboard): the prompts tell a crewmate to compose its page, not pick one - #18758

Merged
chenmingwei23 merged 1 commit into
mainfrom
feat/18642-dashboard-prompt
Oct 12, 2026
Merged

chenmingwei23 merged 1 commit into
mainfrom
feat/18642-dashboard-prompt

Conversation

@chenmingwei23

@chenmingwei23 chenmingwei23 commented Oct 10, 2026 •

Copy link
Copy Markdown
Contributor

Problem / Motivation

Goal: one document tells a conductor how its Dashboard page works -- the dashboard skill -- instead of three prompts describing a flow of their own.

Three prompt sites still describe the view layer the way the old catalog flow worked: an authorized descendant publishes a task-dashboard HTML artifact and updates that slug at each milestone. They are _CONDUCTOR_SYSTEM_PROMPT in agent.py, goal-conductor/SKILL.md and pipeline-conductor/SKILL.md. Two of them also send the reader to the artifacts skill for the page's rules, so the rules live somewhere the dashboard work no longer maintains.

Why it matters

A conductor asked for a board writes a second full board as an artifact while its Dashboard tab already shows one, and rfc-conductor-chat-board-on-tab.md (in-progress on main) measures that cost: "the reader sees every task twice, and each milestone spends output tokens on a full widget plus a task-dashboard artifact update that the tab already shows." A reader who follows the pointer into the artifacts skill then gets page rules that are no longer the ones being developed.

Not a goal

  • No behaviour change. Every edit is prompt text; the only non-prose file is the golden-digest test that measures prompt bytes.
  • What the dashboard skill itself teaches is feat(skills): ship dashboard-manager and retire the template picker #18755's scope, not this PR's.
  • No tool, grant or save flow is added or changed. dashboard_save does not exist on main, and nothing here promises it.
  • The task-dashboard tag, its route and its frame stay exactly as they are.

What changed (motivation -> approach -> change)

Goal: a conductor reads one description of its own dashboard.

Approach. The alternative was to restate the page contract inside each prompt, which is how it drifted into three versions in the first place. Pointing each prompt at the one skill instead means the save and host PRs on #18642 change one document, not four.

Product shape: the decision to stop writing the artifact when the tab is the full board is recorded on main in rfc-conductor-chat-board-on-tab.md, status in-progress, section 3 and its "The task-dashboard artifact" bullet.

  • agent.py, _CONDUCTOR_SYSTEM_PROMPT: the chat widget stays the milestone summary, and the full board is now the Dashboard page, loaded through the dashboard skill. The request for a second task-dashboard artifact at every milestone is gone.
  • goal-conductor/SKILL.md and pipeline-conductor/SKILL.md: each dashboard section opens with the one line the design gives it -- you own your dashboard; a page shows ledger evidence; workers report, never publish -- and defers the page, its flow and its honesty rules to the dashboard skill. Each keeps what is its own: the no-file-writing role and the four non-delegable jobs in the goal conductor, the two columns that remain the source of truth in the pipeline conductor, and in both the rule that no dashboard justifies bypassing a tool approval or raising permission mode.
  • test_agent_refactor_spec_bytes.py: 12 golden digest pairs move, for kirocrew-conductor.json and kirocrew-ledger-conductor.json only.

Backwards compatibility

Compatible: no field, type, validator, tool, route or public function changes. Prompt text and two specs' measured bytes, nothing else.

The diff does take something away -- the instruction to publish a task-dashboard artifact, at three of the four sites that carry it, and the prose pointer into the artifacts skill -- so every reader found by grepping origin/main:

  • Reader: website/src/pages/chat/command-center/useCommandCenter.ts:17 (TASK_DASHBOARD_TAG, behind GET /api/artifacts?tag=task-dashboard) -- entry: crew page, chat -- the tag, the route and the frame are untouched, and artifacts/SKILL.md:58 still teaches artifact_save(name=..., kind="html", tags=["task-dashboard"]), so an artifact that exists still renders. With none, the Overview shows the automatic card, which is already its documented behaviour (docs/feature-map/README.md:295).
  • Reader: website/src/pages/chat/command-center/CommandCenterPanel.tsx:110 and TaskDashboardFrame.tsx:31 -- entry: crew page -- unchanged, behind the PREVIEW_DASHBOARD Feature Preview, default off. website/src/test/commandCenterPanel.test.tsx needs no edit.
  • Reader: src/kiro_crew/builtin_skills/artifacts/SKILL.md:29-58 -- entry: chat, subagent -- the section and the skill's name: are unchanged, so anything that loads it still gets the whole thing. What changes is that the two conductor skills stop pointing at it in prose; the section is left in place because the drop-templates PR on Dynamic Dashboard v3: drop built-in templates, compose a dashboard from a data-type catalog and store it as a dashboard artifact #18642 owns its removal.
  • Reader: docs/system-specs/modules/pipeline-conductor.md:25 -- entry: release -- its Components paragraph still describes the sentence this diff removes, including "published by an authorized descendant", which the new line inverts. docs_lint.py exits 0 because nothing ties the two automatically. Correcting that paragraph is a one-paragraph edit and is the next push on this branch.

Tests

test/test_agent_refactor_spec_bytes.py locks the exact bytes of every packaged agent spec, so any prompt edit that reaches a spec has to be re-measured. 12 digest pairs moved, all of them kirocrew-conductor.json or kirocrew-ledger-conductor.json; no other spec's digest moved and no fileset changed, which is what says the two skills reached only the specs they compose into.

No new test is added: the diff adds no behaviour to assert. The existing guards are what hold it, and they ran on this head with the repository venv, --color=no, -p no:cacheprovider, KIROCREW_HOME at a scratch directory outside the data home.

files result
test_agent_refactor_spec_bytes.py test_skills.py test_conductor_skills_steer_corrections.py test_dashboard_agents_composition_contract.py 283 passed

Gates, each read by its own exit status, all 0: flake8, black --check, check_builtin_skill_scope.py, docs_lint.py, and check_comment_history.py run after its own --test self-test, so the clean result comes from a checker proven able to see a violation.

Manual verification

Both skills were read whole, not diffed alone, to confirm each keeps its own rules; every reader above was grepped on origin/main rather than inferred.

Screenshots / video

N/A -- the diff is prompt text and one test file. Nothing under website/src/ is modified, so there is no rendered delta and no frontend surface changes.

Related Issues

Refs #18642

The T3 one-pager for #18642 was approved, so this line opens its PRs.

The skill these prompts point at is rewritten in #18755.

Checklist

  • At most two commits (one is the norm), with a Conventional Commits title (feat|fix|docs|style|refactor|perf|test|chore|ci|build|revert: ...)
  • 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

Contribution License Agreement

I confirm that this contribution is made under the terms of the Apache-2.0 license, per the Licensing section of CONTRIBUTING.md.

@chenmingwei23
chenmingwei23 requested review from a team and KiroCrewCaptain as code owners October 10, 2026 07:06
@github-actions github-actions Bot added the readiness: checking Automated validation is still running label Oct 10, 2026
@github-actions

github-actions Bot commented Oct 10, 2026 •

Copy link
Copy Markdown
Contributor

UX Review (Opus 5.5) — ⏭️ skipped

Revision e8cdd184b7bf9fb3e9ac4264a6623527914a36b4 touches no user-facing surface (no changes under website/ or committed screenshots), so the UX review was skipped. Advisory — does not block merge.

@github-actions

github-actions Bot commented Oct 10, 2026 •

Copy link
Copy Markdown
Contributor

Design Review (Opus 5.5) — 🟡 CONCERNS

Design-level review of e8cdd184b7bf9fb3e9ac4264a6623527914a36b4 — updated in place on each push. A BLOCK verdict blocks PR readiness; PASS/CONCERNS are advisory.

Design-Verdict: CONCERNS

The description says the per-milestone task-dashboard request "is gone", but the diff keeps it whenever [DASHBOARD] names no board page.

Watch

  • The description and the diff disagree. What changed says "The request for a second task-dashboard artifact at every milestone is gone", and Backwards compatibility says the diff "takes away the instruction… at three of the four sites". The diff keeps it in all three, conditioned on the [DASHBOARD] block: "With no board page named… also keep one task-dashboard artifact and update that same slug". A reader of the description believes the artifact flow is gone from the conductor prompts, when it is now a fallback branch. The later drop-templates PR on Dynamic Dashboard v3: drop built-in templates, compose a dashboard from a data-type catalog and store it as a dashboard artifact #18642 could then miss these three sites when it retires the artifact. The conditional is the safer design; the description just has to say so. Separately, test_agent_refactor_contracts.py's _CONDUCTOR_SYSTEM_PROMPT digest change is not listed in Tests.
    Clears when: the What changed and Backwards compatibility sections describe the artifact as a fallback kept when no board page is named, and Tests lists the contracts-digest move.
  • The fallback branch may never run. dashboard_agentic.turn_block falls back to default_instance(slug) and emits the block whether or not the "Dynamic Dashboard" Feature Preview is on. So a conductor with a default goal-board instance always takes the "page is the full board" branch, even for a user who cannot see that tab. If so, the "no board page named" text is effectively dead, and the full board is invisible to every user with the preview off.
    Clears when: the author confirms which conductor slugs get no default instance (so the fallback is reachable), or states that a preview-off user losing the full board is accepted while both surfaces are preview-gated.

[DESIGN-REVIEWED] e8cdd18

Description read

[DESCRIPTION-READ] 746c07cbefae7f40d52724d2146589d88e14ddb42aaacbd473bc0f4334bcac3a

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 10, 2026 •

Copy link
Copy Markdown
Contributor

First Principles Review (Opus 5.5) — 🟡 CONCERNS

Premise-level review of e8cdd184b7bf9fb3e9ac4264a6623527914a36b4 — 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/CONCERNS are advisory.

First-Principles-Verdict: CONCERNS

The description says the milestone artifact request "is gone", but the diff keeps it at all three sites, and one added clause contradicts the RFC it cites.

Not justified as shipped

  • Item 1 — undeclared: The description says "The request for a second task-dashboard artifact at every milestone is gone" and "No behaviour change". The diff keeps that request at all three sites as a no-board branch ("With no board page named … also keep one task-dashboard artifact"). It also changes what the conductor writes on board tabs, which is a behaviour change.
    Clears when: the description matches the diff: the artifact is conditional at three sites, and the conductor's output changes on board tabs.
  • Item 2 — inherited: The new clause "keep this widget the milestone summary beside it" tells the conductor to keep drawing the full widget on a goal-board/work-kanban tab. The cited RFC table says that branch replies with "'Needs you' lines first … then one line", because "the reader sees every task twice".
    Subtraction: delete "and keep this widget the milestone summary beside it" from _CONDUCTOR_SYSTEM_PROMPT.
    Clears when: the clause is removed, or the RFC's board-tab row is shown to allow the full widget.

What this change ships

Inventory (5 items) — 3 justified

Intent: a conductor on a board tab stops writing a duplicate full-board artifact and reads one dashboard skill. ADDITION, implementing RFC rfc-conductor-chat-board-on-tab (in-progress).

  1. On a goal-board/work-kanban tab, the conductor no longer writes the task-dashboard artifact and loads the dashboard skill — undeclared: the description says the request was removed everywhere
  2. On board tabs, the conductor is told to keep the full widget beside the page — inherited: contradicts the RFC table
  3. The goal-conductor skill now points to the dashboard skill instead of the artifacts skill — justified
  4. The pipeline-conductor skill now points to the dashboard skill instead of the artifacts skill — justified
  5. Golden prompt digests updated for the two conductor specs — justified

[FIRST-PRINCIPLES-REVIEWED] e8cdd18

Description read

[DESCRIPTION-READ] c217c7593718ca4c67e5c5d81dbb50baf4a4a2cbc1d648a2b15d641051640de5

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 10, 2026 •

Copy link
Copy Markdown
Contributor

Opus 5.5 Review — ✅ no blocking findings

Reviewed e8cdd184b7bf9fb3e9ac4264a6623527914a36b4 — this comment is updated in place on each push.

Review details

No findings.

[OPUS-REVIEWED] e8cdd18

Verdict parsed from the review's SHA-scoped output markers for commit e8cdd184b7bf9fb3e9ac4264a6623527914a36b4.

False positive or not applicable? A repository writer can comment:
/ai-review override fable e8cdd184b7bf9fb3e9ac4264a6623527914a36b4: <one-sentence reason>

@github-actions

github-actions Bot commented Oct 10, 2026 •

Copy link
Copy Markdown
Contributor

GPT 6.1 Review — ✅ no blocking findings

GPT 6.1 completed its review of e8cdd184b7bf9fb3e9ac4264a6623527914a36b4 and found no blocking issues.

This comment is updated in place on each push.

Review details

No findings.
[GPT-REVIEWED] e8cdd18

False positive or not applicable? A repository writer can comment:
/ai-review override gpt e8cdd184b7bf9fb3e9ac4264a6623527914a36b4: <one-sentence reason>

@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 10, 2026
@chenmingwei23
chenmingwei23 force-pushed the feat/18642-dashboard-prompt branch from a578a3e to 37605c1 Compare October 10, 2026 07:56
@github-actions github-actions Bot added readiness: checking Automated validation is still running and removed readiness: action required A blocking check or review needs attention labels Oct 10, 2026
@chenmingwei23

Copy link
Copy Markdown
Contributor Author

GPT 6.1 F1 (upheld by adjudication) is fixed at the root on 37605c1b750f58d8ebfdaa1f4381847a02256296, not argued.

The finding was correct. The earlier head replaced the grant-derived universal
assertion with a one-way subset check, so dashboard_templates,
dashboard_preview and dashboard_apply could stay granted while no prompt
said what they were for. That is a loosened ratchet.

What changed:

  • test_dashboard_manager_agent.py now asserts equality in both directions
    in test_the_page_verbs_the_prompt_names_and_the_grants_are_the_same_set.
    Each arm catches a different silent failure: a verb granted and unnamed is a
    tool the crewmate can reach with nothing telling it what the tool is for, and
    a verb named and ungranted sends it to a tool that prompts or does not
    resolve. The verbs are read off _MEMBER_PANEL_GRANTS rather than spelled in
    the test, so a rename on the surface still fails.
  • _MEMBER_DASHBOARD_ITEM therefore names all six granted page verbs. The three
    catalog ones are named in one clause that rules them out of a page request --
    they "compose nothing ... so they answer none of this" -- which satisfies the
    invariant without teaching the flow this PR removes. An explicit exclusion is
    stronger than silence here: a granted tool no prompt mentions is exactly how a
    page gets handed over out of a catalog when somebody asked what the page
    should show.

The adjudicator's other option, removing the three verbs from the grant tuple,
is deliberately not taken: the grant entries and the server's tool definitions
are one pin, and they move together in the PR that removes the tools.

Both arms were shown able to fail before the pass was trusted. Dropping
dashboard_templates from the item reddens it with granted but unnamed;
adding a dashboard_bogus reddens it with named but ungranted. The file was
restored to the same sha256 after each mutation.

Two CI failures on the previous head are fixed in the same push, both real and
both mine:

  • test_builtin_skill_index_line.py -- the author skill's rewritten
    description measured 339 characters against a 300 cap, which truncates its
    Available Skills entry. Trimmed to 295 by dropping the sentence that states a
    rule from inside the skill rather than routing to it.
  • test_agent_refactor_spec_bytes.py -- sixteen golden digests updated, two
    files across eight scenarios, both of them the rewritten charters. The audit
    event sequence, the agent-state sidecar and the file set are unchanged in all
    eight, which is what says the materialization behaviour did not drift and only
    the prompt bytes moved. Each replacement was applied only where the occurrence
    count matched the number of scenarios reporting it, so no other scenario's
    expectation was touched.

1430 tests across four batches on the pushed tree, every command exit 0, plus
flake8, isort, black, mypy, the fold-catalogue gate, docs lint, the
builtin-skill scope gate, and the comment-history gate with its self-test first.

The description is also rewritten so the whole ## Backwards compatibility
reader list now ends inside the 8000-byte description capture, which the Design
lane reported was cutting the list off.

@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 10, 2026
@chenmingwei23
chenmingwei23 force-pushed the feat/18642-dashboard-prompt branch from 37605c1 to 01244b4 Compare October 10, 2026 09:28
@github-actions github-actions Bot added readiness: checking Automated validation is still running readiness: action required A blocking check or review needs attention and removed readiness: action required A blocking check or review needs attention readiness: checking Automated validation is still running labels Oct 10, 2026
@chenmingwei23
chenmingwei23 force-pushed the feat/18642-dashboard-prompt branch from 01244b4 to da29550 Compare October 10, 2026 10:11
@github-actions github-actions Bot added readiness: checking Automated validation is still running readiness: action required A blocking check or review needs attention and removed readiness: action required A blocking check or review needs attention readiness: checking Automated validation is still running labels Oct 10, 2026
@chenmingwei23
chenmingwei23 force-pushed the feat/18642-dashboard-prompt branch from da29550 to ee1fdab Compare October 10, 2026 11:15
@github-actions github-actions Bot added readiness: checking Automated validation is still running and removed readiness: action required A blocking check or review needs attention labels Oct 10, 2026
@github-actions github-actions Bot added readiness: action required A blocking check or review needs attention merge conflict Branch has merge conflicts with its base — author must resolve before merge and removed readiness: checking Automated validation is still running labels Oct 11, 2026
@chenmingwei23
chenmingwei23 force-pushed the feat/18642-dashboard-prompt branch from b3848d1 to ecbf089 Compare October 11, 2026 23:20
@github-actions github-actions Bot added readiness: checking Automated validation is still running and removed readiness: action required A blocking check or review needs attention merge conflict Branch has merge conflicts with its base — author must resolve before merge labels Oct 11, 2026
@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 11, 2026
@chenmingwei23
chenmingwei23 force-pushed the feat/18642-dashboard-prompt branch from ecbf089 to 95afe32 Compare October 11, 2026 23:53
@github-actions github-actions Bot added readiness: checking Automated validation is still running and removed readiness: action required A blocking check or review needs attention labels Oct 11, 2026
@chenmingwei23
chenmingwei23 force-pushed the feat/18642-dashboard-prompt branch from 95afe32 to a0074a5 Compare October 12, 2026 00:54
@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 12, 2026
@chenmingwei23
chenmingwei23 force-pushed the feat/18642-dashboard-prompt branch from a0074a5 to 75a51c4 Compare October 12, 2026 01:25
@github-actions github-actions Bot added readiness: checking Automated validation is still running merge conflict Branch has merge conflicts with its base — author must resolve before merge and removed readiness: action required A blocking check or review needs attention labels Oct 12, 2026
Three prompt sites still described the view layer the way the catalog
flow worked: a descendant publishes a `task-dashboard` HTML artifact and
updates that slug at milestones. With the page now the owner agent's own
HTML carried by the single `dashboard` skill, those sentences send the
reader to the wrong skill for the page's rules.

The rewrite branches exactly where `rfc-conductor-chat-board-on-tab.md`
branches. Its "The `task-dashboard` artifact" bullet keeps the artifact
on the no-tab branch -- "There it is still the only full board" -- so
the artifact is dropped only where the `[DASHBOARD]` block names a board
page (`goal-board` or `work-kanban`), and the request for it stays where
none is named. A reader whose Dashboard tab shows no board therefore
keeps the only full board they have, and no default changes.

`agent.py`'s widget instruction keeps the widget as the milestone
summary and names the Dashboard page as the full board on a board-page
surface, loaded through the `dashboard` skill. The two conductor skills
open their dashboard section with the one line the design gives them --
you own your dashboard page; it shows ledger evidence; workers report to
you and do not publish that page -- defer the page, its flow and its
honesty rules to that skill, and keep the sentence naming who publishes
the artifact for the surface that still needs it. Both keep what is
theirs: the no-file-writing role, the non-delegable jobs, the two
columns of truth, and the rule that no dashboard justifies bypassing a
tool approval or raising permission mode.

The spec-byte goldens move for the two conductor specs those skills
compose into; no other spec's digest or fileset changes.
@chenmingwei23
chenmingwei23 force-pushed the feat/18642-dashboard-prompt branch from 75a51c4 to e8cdd18 Compare October 12, 2026 02:10
@github-actions github-actions Bot removed the merge conflict Branch has merge conflicts with its base — author must resolve before merge label Oct 12, 2026

@buluoray buluoray left a comment

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

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

Approve at e8cdd184b7.

What this brings users: a conductor whose Dashboard page is already the full board (goal-board / work-kanban) stops also writing a second full board as a task-dashboard artifact at every milestone, so the reader no longer sees every task twice and each milestone stops paying for that duplicate. Conductors without such a page keep today's artifact behaviour.

Verified against source

  • The branch only fires where a board page really exists. [DASHBOARD] comes from dashboard_agentic.turn_block (src/kiro_crew/dashboard_agentic.py:722-754), injected only into a crewmate's member block (src/kiro_crew/context_assembly/member.py:544). It names the adopted template, or the default template when nothing was adopted (agent_panel.py:129, dashboard_templates/instance.py:450). So goal-board / work-kanban are named only for a crewmate that adopted one; a plain session or an unadopted crewmate gets no board name and keeps writing the artifact, as before.
  • Nothing a reader relied on is lost. No panel reads a task-dashboard artifact on main (docs/system-specs/modules/pipeline-conductor.md:25, artifacts.md:136), and with the Feature Preview off the tab draws the panel_publish record instead (builtin_skills/dashboard/SKILL.md), which the goal-conductor skill still tells the conductor to publish.
  • Prompt-only: the other changes are the golden digests for the changed prompt bytes. The dashboard skill is bundled under builtin_skills/, so the pointer resolves on every install.

CI on this head: 85 success, 13 skipped, 0 failure. I did not run tests locally; this is a prompt change, checked against source.

Non-blocking follow-ups

  • Until #18755 lands, the dashboard skill teaches how to change the page (search, preview, ask, apply), not "the page, its flow and its honesty rules" these prompts promise. A conductor following the pointer today gets the switching flow.
  • [DASHBOARD] names the board even when the reader has the Feature Preview off, so "that page is the full board" can point at a page the reader cannot see. The panel_publish board covers it, but saying so in the conductor prompt would avoid a conductor telling the user to "open the Dashboard page".

@chenmingwei23
chenmingwei23 enabled auto-merge (squash) October 12, 2026 02:50
@chenmingwei23
chenmingwei23 merged commit 440dca7 into main Oct 12, 2026
101 of 102 checks passed
@chenmingwei23
chenmingwei23 deleted the feat/18642-dashboard-prompt branch October 12, 2026 02:50
@github-actions github-actions Bot removed the readiness: checking Automated validation is still running label Oct 12, 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.

2 participants