mirror of
https://github.com/paperclipai/paperclip.git
synced 2026-10-26 02:00:52 +01:00
9445f4fbbe248a1547ca665319d4e40157df3a73
25
Commits
| Author | SHA1 | Message | Date | |
|---|---|---|---|---|
 scotttongandClaude Fable 5
|
9445f4fbbe |
feat(ui): interstitial polish — hover only on gerund line, no leading icon, held line slides away after dwell
- Expand button moves inside TaskChatStatusPill wrapping only the status line (new onToggle prop); the interstitial row sits outside the hover/ click target - Drop the Responding icon prepending the interstitial text (lead spacer keeps column alignment) - useHeldSelfTalk: a finished update now slides out of the 1lh viewport after --motion-interstitial-dwell (~4s) with no successor, instead of sitting until pushed out; same .tc-line-scroll transition, unmounts to the reserved empty slot Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> |
||
|
scotttongandClaude Fable 5
|
04397ac526 |
fix(ui): insert comment-less settled turns chronologically at run start — assembleThreadItems replaces bottom-append of unanchored Worked rows
A terminal run with no UI-linked reply comment fell into the unanchored bucket and out.push(...unanchored) pinned its "Worked · Ns · N tools" row under every new user bubble. Unanchored turns now interleave into the backbone at meta.startedAt ?? createdAt (Infinity → tail when no meta yet); anchored turns and live-turn-last ordering unchanged. 5 regression tests. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> |
||
|
scotttongandClaude Fable 5
|
eef40c0347 |
feat(ui): round-9 polish — reserved interstitial slot + Worked summary attached to the reply timestamp
The interstitial row above the gerund status line is now permanently reserved while a turn is live: text slides in/out of the always-mounted 1lh slot, so the layout above never jumps as messages appear/disappear. On settle, a turn directly following its own agent's reply bubble attaches to it (attachSettledTurns, after coalesceSettledTurns): the "✓ Worked · …" summary renders appended to the bubble's timestamp line — "2:34 PM · ✓ Worked · 38s · 2 tools" — still expandable to the tool history. Bubble timestamps are always visible (hover-only reveal removed). Turns without a reply bubble keep the standalone folded row. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> |
||
|
scotttongandClaude Fable 5
|
354bb822e0 |
feat(ui): coalesce back-to-back settled agent turns into one "Worked" row — coalesceSettledTurns final pass + buildMergedTurnSummary (summed per-run durations, re-derived counts); user/interaction/live-turn boundaries preserved
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> |
||
|
scotttongandClaude Fable 5
|
39e30b51ab |
feat(ui): interstitial updates on the live line — F10 line-scroll revived (LiveSelfTalkLine, .tc-line-scroll, --motion-line-scroll), ephemeral messages, settled turns slim to tool rows only (thinking rows removed, TaskChatThinkingBlock deleted)
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> |
||
|
scotttongandClaude Fable 5
|
ad1ed3cca3 |
feat(ui): mobile document-flow layout for chat-style task thread — thread renders via scroll={false} in the page scroll with window auto-follow, sticky safe-area composer, 16px mobile editor text; desktop bounded chain untouched
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> |
||
|
scotttongandClaude Fable 5
|
b3c082735c |
feat(ui): ambient self-talk — narration signals via status-word shimmer + Responding icon swap, interstitial text fully suppressed; revert F10 line-scroll and settled nesting, retire P5 interleave
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> |
||
|
scotttongandClaude Fable 5
|
7eadf663ad |
feat(ui): self-talk as a live element of the parent row — streaming text takes the line with 1lh line-scroll, finished self-talk nests like tool rows, settled turns follow
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> |
||
|
scotttongandClaude Fable 5
|
2714832c3d |
feat(ui): task-chat de-emphasized self-talk — bubble-less muted narration with CSS typewriter entry
Mid-run agent messages are tagged interstitial by the transcript adapter and render bubble-less/avatar-less in muted text; the run's posted reply keeps its bubble. Settled history keeps the self-talk interleaved at its streamed timestamp (deduped against posted comment bodies). New --motion-typewriter-* tokens drive a CSS-only steps() clip reveal per streamed block, zeroed under prefers-reduced-motion. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> |
||
|
scotttongandClaude Fable 5
|
76770ace54 |
feat(ui): task-chat live turn collapses to one expandable parent row owning all activity
While a run streams, the status line (whimsy/taxonomy gerund + elapsed + tokens) is now the live turn's single visible row: TaskChatTurnItem gains liveStatus, TaskChatTurn renders it as an expandable header (chevron, same grammar as tool rows / the settled Worked line), and expanding nests the chronological tool/thinking history via the existing .tc-turn-fold motion with live append while open. On settle the header morphs in place into the 'Worked · …' summary, preserving expand state. Only tool/thinking/usage rows nest (isNestableLiveChild); mid-run agent text, markers and interaction cards stay in the thread. Retires the liveTurnFoldedIds window + tests. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> |
||
|
scotttongandClaude Fable 5
|
8bbef4177a |
feat(ui): task-chat composer swaps textarea for shared MarkdownEditor
Rich composer inside the v7 comp-bar chrome: @-mentions (wired through TaskChatThread, previously dropped), data-driven /-command autocomplete via EditorAutocompleteContext, rich lists, and inline pasted/dropped images at the caret through the editor's image plugin -> issue attachments POST. Non-image files keep the chip row; the round-13 object-URL thumbnail path is superseded. Cmd/Ctrl+Enter posts natively; Shift+Tab mode-cycling is captured on the wrapper ahead of Lexical. Motion-gate comment fix in TaskChatStatusPill (doc said 100ms). Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> |
||
|
scotttongandClaude Fable 5
|
15eb30bb64 |
feat(ui): task-chat collapsed turn summary lands below the reply bubble
Settled run turns now anchor after the run's last comment instead of before its first, so the folded "Worked · …" line appears where the live "Running…" pill sat — the pill reads as being replaced by the summary in place. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> |
||
|
scotttongandClaude Fable 5
|
ced262a0da |
feat(ui): task-chat tool taxonomy — pill verbs/icons, Shift+Enter to send, baseline-aligned elapsed
- new shared tool-taxonomy.ts (family/icon/verb per raw ACP tool name,
mcp__ prefix collapse) used by the status pill, chat tool rows, and the
classic transcript's previously hardcoded Wrench sites
- deriveRunStatusLabel: tail tool_call now yields the taxonomy verb
("Searching", "Running a command") with "Tool · target" detail and
the raw toolName so the pill shows the family icon beside the pulse dot
- live status line: label/elapsed/detail/tokens share one baseline row
- TaskChatComposer: Shift+Enter posts, plain Enter inserts a newline
- tests: tool-taxonomy unit tests, deriveRunStatusLabel/toolDisplayName
tests, composer Enter-behavior tests updated
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
|
||
|
scotttongandClaude Fable 5
|
67e0713a2a |
feat(ui): task-chat live status elapsed sits beside the label and ticks in tenths
The running/working status line now shows elapsed time directly to the right of the label (was pushed to the far right edge) and renders it with one decimal (24.3s) so the readout visibly moves. The pill ticks itself every 100ms from the run's start timestamp (new startedAtMs on TaskChatStatusItem), replacing the 1s whole-thread re-render tick in TaskChatThread. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> |
||
|
scotttongandClaude Fable 5
|
61aae371bd |
feat(ui): task-chat per-request mode chips — drop the header mode badge, tag each agent reply with the mode its request ran under
The header's Ask/Plan badge read as a task-global setting, but mode is a per-request choice: with the redesign flag ON the badge is gone from the issue header, and each agent reply's avatar row now shows the work mode that was in effect when its run started, reconstructed from the activity feed's workMode change history (issue.updated details + _previous). Flag OFF is unchanged: the legacy header badge stays and the new workModeChanges prop is ignored by IssueChatThread. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> |
||
|
scotttongandClaude Fable 5
|
9a5f7e26e4 |
feat(ui): task-chat phase 2e — pane tabs in header bar, chat-only center, non-sticky header, plan CTA action bar
Flag-gated (enableTaskChatRedesign), flag OFF byte-identical: - Properties | Plan | Artifacts tab strip portals from IssueProperties into the pane's header bar (new slot left of the maximize/close controls). - Center Chat/Activity/Related-work tab strip hidden — the thread is the page; deep links that switched tabs stay on chat. - Issue header (parent breadcrumb, title/badge block, plugin toolbars, load- earlier button) renders INSIDE the thread scroll viewport, so it scrolls away with the messages instead of staying pinned. - --tc-thread-max-h reserve 21rem → 8rem: composer now sits ~26px above the viewport bottom. - New IssuePlanConfirmationActionBar: while a plan confirmation is pending, its Confirm/Decline CTAs (with decline-reason flow) pin to a footer slot below the pane's scroll area on the Plan tab. Verified: ui tsc clean; motion gate (25 files) + token gates clean; 218 tests across task-chat/pane/IssueDetail/IssueChatThread (1 pre-existing TZ flake in IssueProperties monitor row, fails on base); Playwright on :3093 light+dark — tabs in pane header, header scrolls away/returns, thread bottom 923/950, plan bar pinned at pane bottom in footer slot, zero page errors; flag-off run shows legacy tabs/pane title, 0 redesign testids. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> |
||
|
scotttongandClaude Fable 5
|
a5bdb851ea |
feat(ui): task-chat phase 2c — center/widen shell, inline interaction cards, plan doc in Plan tab
All flag-gated behind enableTaskChatRedesign; flag OFF stays byte-identical. - --tc-shell-max-w (60rem) token; IssueDetail wrapper + skeleton become mx-auto w-full max-w-(--tc-shell-max-w) when the flag is on, and the thread body + pinned composer caps switch to the same token - interaction kind added to the task-chat model; TaskChatThread merges interactions chronologically (same sort keys + same-run handoff shift as buildIssueChatMessages) and renders them via TaskChatInteractionCard, a v7-grammar wrapper over the shared IssueThreadInteractionCard (expired confirmations demote to a marker row) - useIssuePlanDocument hook (404 → null) shared by the Plan tab, which now renders the plan document via MarkdownBody with a revision line above the decompositions, and by the thread's Plan updated/created marker Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> |
||
|
scotttongandClaude Fable 5
|
e76584b95b |
feat(ui): task-chat phase 2b — v7 pixel parity + live streaming
Close the gaps between the flagged redesign and the accepted v7 prototype,
verified against the v7 artifact side-by-side and a REAL streaming agent run:
- Bubbles: human = --liveness-blue #2563eb (was bg-primary near-black) white
text right; agent comment = borderless gray --bubble-agent bubble (new token,
light+dark) with v7 header (capsule avatar · bold name · one capitalized
mode chip). Bubble bodies now render markdown via MarkdownBody (code chips,
links, mentions) instead of raw text.
- Composer: per-mode v7 placeholders ("Message {agent} — describe what you
want done…"), neutral focus-within ring, borderless + attach iconbtn,
full "Agent/Plan/Ask mode" label on the status-chip trigger, icon-only
32px ArrowUp send with hover scale.
- Full-page chat shell: with enableTaskChatRedesign ON, IssueDetail gates the
legacy title (compact single-line editor instead) / description / sub-tasks
table / plan decompositions / Documents (pinned plan) / Output / attachments
wall / workspace card / artifact chips — thread owns the center column,
bounded by new --tc-thread-max-h so TaskMessageScroller owns scrolling with
the composer pinned (scroll viewport = absolute inset-0; percentage heights
don't resolve against flex-sized block parents). Flag OFF proven legacy:
0 redesign testids, all legacy sections present.
- Tool rows (v7 flat grammar): [glyph · name · mono target · hover chevron ·
status]; transcript adapter now emits real tool names (toolDisplayName —
mcp__ collapse + capitalization, tool_result.toolName backfill), summarized
primary params (summarizeToolInput probes file_path/command/pattern/…),
clipped mono result detail, generic "Tool" fallback for legacy stored logs.
- Streaming states: shimmer (.shimmer-text-muted variant) on live Thinking…
header + flat pulse-dot statusline (running/working no longer card chrome);
thinking settles to "Thought for Ns" (ts-span derived) and arrives collapsed
in settled history; turn summary label + mono metrics per v7 runsum.
Verified: motion gate (24 files, seams), token gates, workspace tsc, 177 ui
tests incl. task-chat/PropertiesPanel/IssueChatThread/IssueDetail; live
streaming captured on a real claude_local run on the worktree instance
(:3093 scratch task) — thinking/text deltas, live statusline with elapsed,
auto-follow, and fold-to-summary on completion; flag-off isolation screenshot.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
|
||
|
scotttongandClaude Fable 5
|
474c4138ac |
feat(ui): task-chat redesign phase 2 — v5–v7 reconciliation + live-thread completeness
Reconcile the baseline with the accepted v5–v7 prototype decisions and finish the live-thread work items (, plan §5), all behind enableTaskChatRedesign: - Composer: v7 comp-bar (attach + far left, rectangle status-chip mode trigger via --tc-mode-* hues incl. ask-blue #3b82f6, assignee combobox left of send, all controls 32px, no shortcut hints), per-message mode with ⇧⇥ cycling, attachments via button/paste/drag-drop, reassign-on-send with implicit reopen. - Scroll pill: stock shadcn MessageScrollerButton (32px round, arrow-only) with tokenized entry/exit motion and easing-aware re-follow (smooth glide cannot unpin the follow state; wheel/touch cancels). - Turn fold: run activity grouped into TaskChatTurn — live turns stream expanded, settling folds to a one-line '✓ Worked · 38s · 3 tools · +34 −3' summary (--motion-turn-fold 380ms, instant for settled history, reduced-motion honored), settled turns interleave before their run's reply comment. - Bubbles: agent header gains the textless brand-gradient capsule avatar + name · mode chip (deterministic agent→gradient hash). - Properties pane: tab label Plans→Plan; drag-resizable (clamp ≥260px, dbl-click reset 322px, width remembered) + maximizable (320ms glide, spacer holds flex slot, 840px content cap) — flag-off DOM untouched. - Motion: new tokens (turn-fold, scroll-pill enter/exit + easings, pane-glide) in index.css :root + reduced-motion zeroing + tweak-panel catalog entries. Verified: workspace tsc clean; check-task-chat-motion + token gates clean; 53 task-chat/PropertiesPanel/gate tests, 29 experimental-settings tests and 21 server instance-settings tests pass. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> |
||
|
scotttongandClaude Fable 5
|
d270ee47fd |
feat(ui): task-chat redesign baseline behind enableTaskChatRedesign
New ui/src/components/task-chat/ surface (model, TASK_CHAT_STATES inventory, fixtures, adapter, Bubble/Marker/StatusPill/ToolCard/ThinkingBlock/UsageReadout/ Scroller/PlanView/ThreadView, TweakPanel), motion tokens in index.css :root, useTaskChatRedesignEnabled hook + TaskChatRedesignGate, one-line ThreadComponent seam in IssueDetail.tsx, flag-scoped tabs in IssueProperties.tsx, dev harness /dev/task-chat-lab, gate script scripts/check-task-chat-motion.mjs. Flag wired through the 3 lockstep shared files (types/validators/feature-catalog), preference tier, default off. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> |
||
|
Nicky LeachandPaperclip
|
01b51dc0e5 |
fix(runtime): stop Live badge and Working shimmer after task teardown (#10985)
## Thinking Path > - Paperclip is the open source app people use to manage AI agents for work. > - The task list and the chat views show a Live badge and a Working shimmer for an issue that has an active run. > - A finished task kept the Live badge and the Working shimmer after the run ended and the sandbox stopped. > - The user interface reads run liveness from the `heartbeat_runs.status` row. The run finalizer writes the terminal status in a step that is separate from the agent `status=done` update. When the sandbox or the run process stops between the two steps, `heartbeat_runs.status` stays `running` forever. > - A run row that stays `running` makes a finished task look perpetually Live, and the user interface has no guard for an issue that already reached a terminal status. > - This pull request closes the invariant "environment lease released implies the run is terminal" on the server, and adds a user interface guard that suppresses live state for a terminal issue. > - The benefit is that a finished task stops showing Live and Working, both at the source (the run row) and at the surface (the badge and the shimmer). ## Linked Issues or Issue Description **Bug description** - A completed task kept the Live badge and the Working shimmer after its run ended and the sandbox was torn down. **Steps to reproduce** - Run an agent task to completion. Let the sandbox tear down while the run finalizer is between the `status=done` update and the terminal run-status write. - Open the task list or the chat view for the finished task. **Expected behavior** - A finished task shows no Live badge and no Working shimmer. **Actual behavior (before this change)** - The finished task showed the Live badge and the Working shimmer because its `heartbeat_runs.status` row stayed `running`. This pull request supersedes the two separate pull requests #10954 (frontend) and #10955 (backend). It carries all of their changes for the same race. ## What Changed Server: - Run teardown terminalizes a still-running or still-queued run before it releases the environment lease. It writes `succeeded` when the issue already reached `done`, `cancelled` when the issue is `cancelled`, and `interrupted` otherwise. It never overwrites a status that another path already made terminal. - The recovery stale-lock sweep terminalizes an orphaned running run to `interrupted` after it confirms the process and the sandbox are both gone. It requires recorded process metadata, so it never terminalizes a live run, a queued run, or a scheduled retry. - Each terminal transition writes a run event. - The stale-lock sweep continues and clears the lock when the audit write fails. It logs the failure loudly. - New server tests cover both invariants. User interface: - A shared guard suppresses the Live badge and the Working shimmer when the issue status is terminal. - The guard keeps non-terminal `queued` and `running` issues live. - The guard prefers the newest issue live-status snapshot. - New user interface tests cover the guard and the snapshot preference. ## Verification Server: - `pnpm --filter @paperclipai/plugin-sdk ensure-build-deps && tsc --noEmit` in `server/` — 0 errors. - `pnpm --filter server test heartbeat-run-lease-release-terminalization.test.ts recovery-stale-issue-lock-sweep.test.ts` — 12 tests pass. User interface: - `pnpm --filter @paperclipai/ui typecheck` — 0 errors. - `pnpm exec vitest run ui/src/lib/liveIssueIds.test.ts ui/src/lib/issue-chat-messages.test.ts` — 40 tests pass. ## Risks - Low risk. The server change only forces a still-live run row to a terminal status when the lease releases or when the recovery sweep confirms the process is dead. It never overwrites an existing terminal status, and it guards the recovery path with process metadata to avoid terminalizing a live run. - The user interface change is additive. The guard only suppresses live state for a terminal issue and keeps queued and running issues live. - No database migration. No change to any external endpoint. ## Model Used - Claude Opus 4.8 (`claude-opus-4-8`), extended thinking, tool use, and code execution. ## Checklist - [x] I have included a thinking path that traces from project context to this change - [x] I have specified the model used (with version and capability details) - [x] I have checked ROADMAP.md and confirmed this PR does not duplicate planned core work - [x] I have searched GitHub for duplicate or related PRs and linked them above - [x] I have either (a) linked existing issues with `Fixes: #` / `Closes: #` / `Refs: #` OR (b) described the issue in-PR following the relevant issue template - [x] I have not referenced internal/instance-local Paperclip issues or links (only public GitHub `#NNN` / `github.com/paperclipai/paperclip` URLs) - [x] My branch name describes the change (e.g. `docs/...`, `fix/...`) and contains no internal Paperclip ticket id or instance-derived details - [x] I have run tests locally and they pass - [x] I have added or updated tests where applicable - [x] I have updated relevant documentation to reflect my changes - [x] I have considered and documented any risks above - [x] All Paperclip CI gates are green - [x] Greptile is 5/5 with no open P2s, recommendations, or follow-ups - [x] I will address all Greptile and reviewer comments before requesting merge --------- Co-authored-by: Paperclip <noreply@paperclip.ing> |
||
|
scotttongandClaude Opus 4.8
|
ea83c5c822 |
feat(task-chat): bring back copy/👍/👎 actions on the agent bubble footer (#11025)
## Thinking Path > - Paperclip is the open source app people use to manage AI agents for work. > - The task view shows a threaded conversation between the user and the agent, with agent replies rendered as bubbles that carry a "✓ Worked · N tools" summary line. > - The conference-room chat already offers per-message copy and thumbs-up / thumbs-down feedback, but the redesigned task thread dropped these controls from the agent bubble footer. > - Users lose a quick way to copy an agent reply or send feedback on it, and the redesign silently ignored the feedback-vote props it was already given. > - This pull request prepends a copy · thumbs-up · thumbs-down cluster to the bubble summary line and wires the existing feedback-vote props through. > - The benefit is a consistent feedback surface across both chat views, with no new API. ## Linked Issues or Issue Description No public GitHub issue exists for this change; the underlying issue is described inline below following the feature template (`.github/ISSUE_TEMPLATE/feature_request.yml`). #### Problem or motivation The redesigned task thread renders each agent reply with a "✓ Worked · N tools · <timestamp>" summary line, but it dropped the copy and thumbs-up / thumbs-down controls that the conference-room chat still shows. Users can no longer copy an agent reply or vote feedback from the task thread. The redesign component already received `feedbackVotes` and `onVote` props but ignored them. #### Proposed solution Prepend a copy · 👍 · 👎 cluster to the summary line, leading the always-visible timestamp, reusing the shared `IssueChatFeedbackButtons` so both chat views speak the same feedback language. Anchor the cluster to the turn's summary row (a sibling of the expandable tool-history fold) so it stays on the summary line whether the tool history is collapsed or expanded. #### Alternatives considered Placing the cluster inside the expandable fold — rejected because expanding the tool history then re-centered the actions to the middle of the tall fold. #### Roadmap alignment UI polish to the task thread; no core-roadmap overlap. ## What Changed - Add `TaskChatBubbleActions`: a copy · thumbs-up · thumbs-down cluster built on the shared `IssueChatFeedbackButtons`. - Render the cluster on the agent bubble's "✓ Worked · …" summary line, leading the timestamp; runless agent replies get the same cluster with the timestamp trailing. Human and system bubbles are unchanged. - Add a `leading` slot to `TaskChatTurn` so the actions sit on the summary row, a sibling of the tool-history fold, and stay anchored when the fold expands. - Wire the redesign to the `feedbackVotes` / `onVote` props it already received. - Add a demo binding in the `TaskChatLab` dev harness. ## Verification - `pnpm check:token-gates` — 3/3 gates CLEAN. - `pnpm typecheck` — clean across all packages. - `cd ui && pnpm vitest run src/components/task-chat/TaskChatBubble.test.tsx src/components/task-chat/TaskChatTurn.test.tsx` — 31/31 pass. - Manual: open a task thread, confirm the copy / 👍 / 👎 cluster shows on the agent bubble summary line before the timestamp, copy works, votes toggle, and the cluster stays on the summary line when the tool history is expanded. Visual change: snapshot baselines are intentionally not updated, per the `doc/design/DECISION-SHEET.md` entry "Per-change snapshot verification demoted to dormant (Jul 13 2026)". ## Risks Low risk. UI-only change scoped to the redesigned task-chat bubble footer. It reuses an existing shared feedback component and existing vote props; no API, schema, or server change. Human and system bubbles are untouched. ## Model Used Claude Opus 4.8 (Anthropic), model id `claude-opus-4-8`, extended thinking enabled, tool use. ## Checklist - [x] I have included a thinking path that traces from project context to this change - [x] I have specified the model used (with version and capability details) - [x] I have checked ROADMAP.md and confirmed this PR does not duplicate planned core work - [x] I have searched GitHub for duplicate or related PRs and linked them above - [x] I have either (a) linked existing issues with `Fixes: #` / `Closes #` / `Refs #` OR (b) described the issue in-PR following the relevant issue template - [x] I have not referenced internal/instance-local Paperclip issues or links (only public GitHub `#NNN` / `github.com/paperclipai/paperclip` URLs) - [x] My branch name describes the change (e.g. `docs/...`, `fix/...`) and contains no internal Paperclip ticket id or instance-derived details - [x] I have run tests locally and they pass - [x] I have added or updated tests where applicable - [x] I have updated relevant documentation to reflect my changes - [x] I have considered and documented any risks above - [x] All Paperclip CI gates are green - [x] Greptile is 5/5 with no open P2s, recommendations, or follow-ups - [x] I will address all Greptile and reviewer comments before requesting merge --------- Co-authored-by: Claude Opus 4.8 <noreply@anthropic.com> |
||
|
DottaandClaude Opus 5
|
e8ae5286eb |
feat(issues): explain cross-task agent writes with attribution, audit receipts, and actionable denials (#10843)
## Thinking Path
> - Paperclip is the open source app people use to manage AI agents for
work
> - Agents write to tasks they do not own. They comment, they change
fields, and the control plane now permits this by default for
standard-trust agents on any task they can read
> - This makes a task thread ambiguous. A reader sees a comment from an
agent that is not the assignee, but no surface says whose authority that
write rode
> - The same gap applies to field edits. The activity stream named the
verb, but it did not show the before value, the after value, or the
reason the write was permitted
> - The remaining refusals are also opaque. An agent that hits a wall
receives a 403 with no boundary name, no actor who can act, and no
sanctioned path. One real incident spent a full detour to find the
workaround
> - This pull request adds the three surfaces that make open cross-task
writes legible: an attribution chip, a field-level audit receipt, and an
actionable denial contract shared by the API and the UI
> - The benefit is that a reader can answer "who did this, on whose
authority, and was it allowed?" on the task itself, and a blocked writer
is told what to do next
## Linked Issues or Issue Description
No public issue exists for this work, so the enhancement is described
here.
**What existing behavior does this improve?**
Cross-task agent writes are permitted, but they are not explained. A
task thread can hold comments from agents that are not the assignee, and
the activity stream can hold field changes made by those agents. Neither
surface names the responsible user behind the write. When a write is
refused, the error text does not name the boundary or the way forward.
**Subsystem affected**
Issue detail UI (comment thread and activity stream), the issue write
authorization responses in the server, and the shared copy contract that
both consume.
**Current behavior**
- An agent comment on a task the agent does not own looks the same as an
assignee comment.
- An `issue.updated` activity row states the verb only. It does not show
the field-level before and after values, the responsible user, or the
authorization reason.
- A refused write returns a short message such as an ownership error.
The message does not state which rule fired, who is able to perform the
action, or which alternative path is sanctioned.
**Proposed behavior**
- An agent comment on a task the agent does not own carries a chip that
reads "for {user}". The chip names the responsible user. Its tooltip
states that the author is not the assignee and cannot exceed that user's
permissions.
- Each `issue.updated` row shows a receipt: the changed fields with
before and after values, the responsible user, and the authorization
reason. This applies to board edits as well as agent edits.
- Each refusal states three things: the boundary that fired, who is able
to act, and the sanctioned path. The API error body and the in-app
notice use the same words, because both read one shared contract.
Related pull requests, found by searching this repository:
- Refs #10837 — merged. It added the default-open cross-task write rule,
the comment attribution data, and the per-run containment cap that this
pull request makes visible.
- Refs #10114 — open. It proposes a narrower authorization change in the
same area.
- Refs #7998 — open. It proposes append-only cross-assignee comments as
an alternative to opening writes.
## What Changed
- Adds `packages/shared/src/issue-write-denial.ts`. This is one copy
contract for eight ways an issue write can be refused: not visible,
responsible-user ceiling, responsible user unavailable, excluded actor
class, assignee run lock, per-run cross-task cap, missing run context,
and rejected attribution. Each entry names the boundary, who can act,
and the sanctioned path.
- Maps server authorization decisions onto that contract in
`server/src/routes/issues.ts` and
`server/src/services/cross-issue-influence-limit.ts`. The flattened
`error` string carries all three obligations, and `details.code` lets
the UI render the same words. The two cap codes keep the names they
already ship under.
- Adds `CommentAttributionChip`. It renders "for {user}" beside the
author name on agent comments where the author is not the assignee. It
renders nothing when no responsible user is recorded, so older rows stay
clean. It is wired into both `IssueChatThread` and the flagged
`TaskChatThread` redesign.
- Adds `IssueFieldChangeReceipt`. It renders the change receipt under
`issue.updated` rows in the activity stream. Ids resolve to agent and
user names where the directory is loaded. Server-truncated text is
labelled as a preview, so the receipt never implies that it shows a
whole value.
- Adds `IssueWriteDenialNotice`. It renders the shared copy in the app,
keyed off the denial events the server logs on a task.
- Adds a public `/ux-lab/cross-issue-collaboration` page. It renders all
three surfaces and their edge cases for review without a seeded thread.
This follows the existing `ux-lab` pages.
## Verification
Automated, all green:
```
pnpm --filter @paperclipai/shared exec vitest run src/issue-write-denial.test.ts # 17 tests
pnpm --filter @paperclipai/ui exec vitest run src/components/IssueWriteDenialNotice.test.tsx \
src/components/IssueFieldChangeReceipt.test.tsx src/components/CommentAttributionChip.test.tsx \
src/lib/issue-change-receipt.test.ts src/lib/comment-attribution.test.ts # 46 tests
pnpm --filter @paperclipai/server exec vitest run src/__tests__/cross-issue-influence-limit.test.ts \
src/__tests__/issue-comment-attribution-audit-routes.test.ts \
src/__tests__/issue-agent-mutation-ownership-routes.test.ts \
src/__tests__/low-trust-red-team-routes.test.ts # 98 tests
```
`tsc --noEmit` passes for the shared, ui, and server packages.
Manual, in a browser:
1. Start the UI only: `pnpm --filter @paperclipai/ui exec vite`.
2. Open `/ux-lab/cross-issue-collaboration`. No session is needed,
because `ux-lab` routes are public.
3. All three surfaces were captured at 1440x900 in light mode and dark
mode, and at 390x844. The page reported no errors.
4. The chip tooltip was opened by a hover and by a keyboard focus.
Rendering the page found defects that the tests had missed. Three copy
and contrast defects were fixed, and two of them are now pinned by a
test. A design review then found three layout defects, which are also
fixed: the denial notice orphaned its label when a value wrapped, the
receipt icon wrapped onto its own line at narrow widths, and the chip
tooltip was reachable by hover only.
## Risks
Low risk, and additive.
- Every new surface renders nothing when its data is absent. Comments
without a recorded responsible user show no chip, and activity events
without a receipt show no receipt, so existing rows do not change.
- No migration is included. The data these surfaces read already ships.
- The wire values of the two per-run cap denial codes are unchanged.
Only the human-readable text changes, plus six codes that had no
`details.code` before.
- The denial copy is read by agents as well as people. If wording must
change later, one shared module is the only place to change it.
- Roadmap check: this extends the completed "Activity log & action
attribution" area rather than duplicating planned core work.
## Model Used
Claude Opus 5 (Anthropic), model id `claude-opus-5[1m]`, 1M context
window, extended thinking, with tool use and code execution. It ran as
an agent in Claude Code and drove a real browser to capture the review
screenshots.
## Checklist
- [x] I have included a thinking path that traces from project context
to this change
- [x] I have specified the model used (with version and capability
details)
- [x] I have checked ROADMAP.md and confirmed this PR does not duplicate
planned core work
- [x] I have searched GitHub for duplicate or related PRs and linked
them above
- [x] I have either (a) linked existing issues with `Fixes: #` / `Closes
#` / `Refs #` OR (b) described the issue in-PR following the relevant
issue template
- [x] I have not referenced internal/instance-local Paperclip issues or
links (only public GitHub `#NNN` / `github.com/paperclipai/paperclip`
URLs)
- [x] My branch name describes the change (e.g. `docs/...`, `fix/...`)
and contains no internal Paperclip ticket id or instance-derived details
- [x] I have run tests locally and they pass
- [x] I have added or updated tests where applicable
- [x] I have updated relevant documentation to reflect my changes
- [x] I have considered and documented any risks above
- [ ] All Paperclip CI gates are green
- [ ] Greptile is 5/5 with no open P2s, recommendations, or follow-ups
- [x] I will address all Greptile and reviewer comments before
requesting merge
---------
Co-authored-by: Claude Opus 5 (1M context) <noreply@anthropic.com>
|
||
|
|
9b9631b724 |
feat(ui): chat-style tasks polish — rich-text composer, attachment chips, live-turn interstitials, mobile layout (#10707)
## Thinking Path > - Paperclip is the open source app people use to manage AI agents for work. > - Operators talk to their agents on the issue detail page. An experimental "Chat-Style Tasks" view (#10606) makes that page read as a conversation instead of a ticket form. > - The first release of that view shipped with a plain-text composer, no visible narration while an agent works, and a desktop-only layout. > - Users write formatted replies, paste screenshots, and follow long agent runs from their phones. The experimental view should support all of that before it can graduate. > - This pull request is the next iteration of the same experiment: a rich-text composer with attachments, live-turn narration on the status line, cleaner settled-turn history, and a mobile layout. > - The benefit is a chat view that feels alive while the agent works and stays readable after it finishes, on desktop and mobile, still fully behind the existing opt-in flag. ## Linked Issues or Issue Description Refs #49 (chat with agents is a much-wanted feature). Refs #10606 (the merged first release of the experimental chat-style task view; this PR iterates on it). Related PRs found in the dedup search: - #8228 — open PR that polishes the classic issue chat composer. It targets the flag-off legacy path; this PR only changes the flag-on experimental view. - #10466 — merged blockquote-recovery fix in the shared MarkdownEditor. This PR now reuses that editor inside the chat composer. **What existing behavior does this improve?** The experimental "Chat-Style Tasks" view on the issue detail page (Settings → Experimental, `enableTaskChatRedesign`, default off). **Subsystem affected** UI (issue detail page, chat-style task view). **Current behavior** With the experiment enabled, the composer is a plain textarea with no formatting, no attachment preview, and no mention support. While an agent runs, the status line shows only a static label, and the agent's narration text is hidden. Finished runs render one settled row per turn, so a run with many short turns produces a long list of near-duplicate "Worked" rows, and turns without a comment append at the bottom out of order. On mobile, the desktop bounded-height thread makes the page scroll poorly. **Proposed behavior** The composer uses the shared MarkdownEditor: markdown formatting, mentions, image paste with thumbnail previews, and non-image attachment chips. Sending posts on Cmd/Ctrl+Enter. While an agent runs, the status line rotates playful status words and surfaces the agent's own narration as short interstitial updates: each update holds for a minimum dwell, is replaced only when superseded, and slides through a one-line viewport with tokenized motion. Back-to-back settled turns coalesce into one "Worked" row with summed durations and re-derived tool counts, and comment-less settled turns insert chronologically at their run's start time. On mobile, the thread renders in the document flow with window auto-follow and a sticky safe-area composer; the desktop layout is unchanged. **Reason and benefit** The chat view is only convincing if it feels like a conversation with a working agent. Rich text and screenshots are table stakes for chat input. Live narration gives moment-to-moment feedback without opening transcripts. Coalesced history keeps long-running tasks readable. Mobile support lets operators follow runs away from their desks. **Breaking changes** None. Every change is gated behind the existing `enableTaskChatRedesign` flag, which is off by default. The flag-off page is unchanged. ## What Changed - `TaskChatComposer` swaps its textarea for the shared `MarkdownEditor`: markdown formatting, mentions, image paste with object-URL thumbnail previews (revoked on clear and unmount), and posting on Cmd/Ctrl+Enter. - Non-image attachments render as chips on a new shared `ui/attachment.tsx` primitive (adds the `@base-ui/react` dependency it builds on). - New `status-whimsy.ts`: deterministic rotation of playful status words on the live status line. - Live interstitial narration: the transcript adapter tags agent self-talk, and the live status line shows it as ephemeral one-line updates with a ~4s minimum dwell, hold-until-superseded replacement, and a slide transition driven by new `--motion-line-scroll` tokens (cataloged in `motion-tokens.ts`, which a test keeps 1:1 with `index.css`). Hover affordance applies only to the status line, with no leading icon. - Settled-turn history: `coalesceSettledTurns` merges back-to-back settled agent turns into one "Worked" row (summed per-run durations, tool counts re-derived from the merged turn); `assembleThreadItems` inserts comment-less settled turns chronologically at run start instead of appending them at the bottom; settled turns render tool rows only (the separate thinking block component is removed). - The "Worked" summary attaches to the reply timestamp row, and thread timestamps are always visible. - Mobile layout: the thread renders with `scroll={false}` in the page scroll, a new `useWindowAutoFollow` hook keeps the window pinned to new content, the composer is sticky with safe-area padding, and the editor uses 16px text so iOS does not zoom on focus. The desktop bounded chain is untouched. - New `TaskChatDescriptionBubble` renders the issue description as the first chat bubble, and `McpIcon` gives MCP tools a distinct icon. - `IssueDetail.test.tsx` stubs `TaskChatThread`: the composer's `@mdxeditor` dependency cannot load under jsdom's CSSOM, and the suite exercises the flag-off path. ## Verification - `pnpm check:token-gates` — 3/3 CLEAN. - `node scripts/check-task-chat-motion.mjs` — OK (30 files scanned, seams present). - `cd ui && npx tsc -b` — clean. - `cd ui && pnpm vitest run` — 3,482 of 3,483 tests pass locally. The one failure is the `IssueProperties.test.tsx` monitor-row time-formatting test, which is timezone-sensitive: it fails identically on unmodified `origin/master` in a non-UTC timezone and passes with `TZ=UTC`. It is not related to this change. - Manual: enable "Chat-Style Tasks" in Settings → Experimental and open an issue with an assigned agent. Comment to start a run: the status line rotates status words and shows the agent's narration as short held updates. After the run, consecutive turns fold into one "Worked" row under the reply timestamp. Paste an image into the composer to see a thumbnail chip; attach a non-image file to see a file chip; send with Cmd+Enter. Open the same issue in a narrow viewport to see the document-flow layout with the sticky composer. - Visual snapshot baselines are intentionally not updated: per `doc/design/DECISION-SHEET.md`, "Per-change snapshot verification demoted to dormant (Jul 13 2026)". ## Risks - The composer now loads the shared MarkdownEditor inside the chat view. The editor is already used across the app (issue descriptions, comments), so its behavior is well exercised; composer-specific handling (paste, attachments, submit keys) is covered by new tests. - The transcript adapter changes how live narration and settled turns are derived from run logs. Malformed or legacy logs degrade to generic rows rather than crashing, and the adapter suites cover the merge and ordering rules. - Object URLs for paste previews are revoked on send-clear and unmount to avoid leaks; jsdom environments without `URL.createObjectURL` are guarded. - All changes are behind the default-off `enableTaskChatRedesign` flag. Overall risk with the flag off is low. ## Model Used - Claude (Anthropic), model id `claude-fable-5` (Claude Fable 5), extended thinking enabled, agentic tool use (file editing, shell, test execution) via Claude Code / Claude Agent SDK. ## Checklist - [x] I have included a thinking path that traces from project context to this change - [x] I have specified the model used (with version and capability details) - [x] I have checked ROADMAP.md and confirmed this PR does not duplicate planned core work - [x] I have searched GitHub for duplicate or related PRs and linked them above - [x] I have either (a) linked existing issues with `Fixes: #` / `Closes #` / `Refs #` OR (b) described the issue in-PR following the relevant issue template - [x] I have not referenced internal/instance-local Paperclip issues or links (only public GitHub `#NNN` / `github.com/paperclipai/paperclip` URLs) - [x] My branch name describes the change (e.g. `docs/...`, `fix/...`) and contains no internal Paperclip ticket id or instance-derived details - [x] I have run tests locally and they pass - [x] I have added or updated tests where applicable - [x] I have updated relevant documentation to reflect my changes - [x] I have considered and documented any risks above - [x] All Paperclip CI gates are green - [x] Greptile is 5/5 with no open P2s, recommendations, or follow-ups - [x] I will address all Greptile and reviewer comments before requesting merge --------- Co-authored-by: Paperclip <noreply@paperclip.ing> Co-authored-by: Claude Fable 5 <noreply@anthropic.com> |
||
|
scotttongandClaude Fable 5
|
c185e64b77 |
feat(ui): chat-style task view behind an experimental flag (#10606)
## Thinking Path > - Paperclip is the open source app people use to manage AI agents for work. > - Operators spend most of their time on the issue detail page. They talk to the assigned agent there through comments. > - The current page reads as a ticket form. The thread sits below properties, the composer sits mid-page, and live agent activity renders as dense transcript logs. > - Talking to an agent is a conversation. A chat-first layout matches that mental model better than a ticket form. > - A layout change this large must not disrupt current users. It needs a safe opt-in path and full parity with the existing thread features. > - This pull request adds a chat-style task view behind a new "Chat-Style Tasks" experiment toggle. The flag is off by default and the existing page is unchanged when it is off. > - The benefit is a focused, readable conversation with the agent: live tool activity folds into compact summaries, the composer stays at the bottom, and properties, plan, and artifacts move into header tabs. ## Linked Issues or Issue Description Refs #49 (chat with agents is a much-wanted feature). Related PRs found in the dedup search: - #4489 — an earlier, closed attempt to promote the conversation to the primary surface on issue detail. This PR is a fresh, flag-gated take on the same goal. - #8837 — an open PR that proposes a two-column task layout. It restructures the same page but keeps the ticket paradigm; this PR is orthogonal because it is opt-in and chat-first. **Subsystem affected** UI (issue detail page). **Problem or motivation** The issue detail page presents agent conversations as a ticket: properties first, thread below, composer in the middle of the page, and raw transcript noise during live runs. Users who mainly converse with their agents must scroll past chrome to follow the conversation, and live activity is hard to read. **Proposed solution** An opt-in chat-style view of the issue detail page, gated by a new "Chat-Style Tasks" experiment toggle in Settings → Experimental. With the flag on, the thread fills the center pane, the composer docks to the bottom of the viewport, Properties / Plan / Artifacts become header tabs, live turns show a status pill with the current tool action and elapsed time, and settled turns collapse to a "Worked · N tools" summary that expands into per-tool rows. With the flag off, nothing changes. **Alternatives considered** Restyling the existing layout in place (rejected: too disruptive without an opt-out), and a separate chat page beside the issue page (rejected: splits the task's single source of truth). A per-request lab page (`/task-chat-lab`, dev-only) was kept for design iteration instead. **Roadmap alignment** ROADMAP.md "CEO Chat" wants lighter conversations that still resolve to real work objects. This PR keeps the core task-and-comments model — it only changes presentation, opt-in — so it does not duplicate that planned work. ## What Changed - New `enableTaskChatRedesign` instance setting, exposed as a "Chat-Style Tasks" experiment card in Settings → Experimental (shared feature catalog, validators, server instance-settings service, and UI settings page). - New `ui/src/components/task-chat/` component family: chat thread with turn grouping, agent reply bubbles, live status pill, collapsible turn summaries with per-tool rows, plan tab with a sticky CTA action bar, inline interaction cards, per-request mode chips, and a bottom-docked composer. - A shared tool taxonomy (`tool-taxonomy.ts`) maps tool names to verbs and icons; the status pill, tool rows, and the classic transcript view all use it. - A transcript adapter converts stored run logs into chat turns; it dedupes tool-call updates by `toolUseId` so tool counts match the expanded rows, and it keeps a tool row's first real name when later generic updates arrive. - Composer: posts on Cmd/Ctrl+Enter, supports image paste with object-URL thumbnail previews (revoked on clear/unmount), and uploads through the issue attachments route. - `IssueDetail.tsx`: with the flag on, pane tabs move to the header bar, the header is not sticky, and the chat fills the center; with the flag off, the previous layout renders unchanged. - Motion tokens for the new animations live in `ui/src/index.css` with a `motion-tokens.ts` catalog and a test that keeps the two in sync (the catalog now also covers the shared enter/exit/swap tokens that the decision/quicklook block declares). - A dev-only `/task-chat-lab` page with fixtures and a tweak panel for motion tuning. ## Verification - `pnpm typecheck` — clean across the workspace. - `pnpm check:token-gates` — 3/3 CLEAN. - `cd ui && pnpm vitest run` — 3,344 of 3,345 tests pass locally. The one failure is `IssueProperties.test.tsx` monitor-row time formatting, which is timezone-sensitive: it also fails on unmodified `origin/master` in a non-UTC timezone and passes with `TZ=UTC`. It is not related to this change. - `cd server && pnpm vitest run src/__tests__/instance-settings-service.test.ts` — 21/21 pass (covers the new setting). - Manual: start the dev server, open Settings → Experimental, enable "Chat-Style Tasks", and open any issue. The thread fills the page, the composer docks to the bottom, and Properties / Plan / Artifacts appear as header tabs. Assign an agent and comment to watch a live run: the status pill shows the current tool action with elapsed time, and the finished turn folds into a "Worked · N tools" summary. Disable the toggle and confirm the classic page is unchanged. - Visual snapshot baselines are intentionally not updated: per `doc/design/DECISION-SHEET.md`, "Per-change snapshot verification demoted to dormant (Jul 13 2026)". ## Risks - The flag-off path goes through the same `IssueDetail.tsx` file, so a regression there would affect current users. Mitigation: the classic markup renders through the same components as before behind explicit flag conditionals, and the full UI suite passes. - The transcript adapter interprets stored run-log formats, including legacy entries without `toolUseId`. Malformed logs degrade to generic tool rows rather than crashing. - The new view changes no server behavior other than one additive instance setting; it is additive and default-off. Overall risk with the flag off is low. ## Model Used - Claude (Anthropic), model id `claude-fable-5` (Claude Fable 5), extended thinking enabled, agentic tool use (file editing, shell, test execution) via Claude Code / Claude Agent SDK. ## Checklist - [x] I have included a thinking path that traces from project context to this change - [x] I have specified the model used (with version and capability details) - [x] I have checked ROADMAP.md and confirmed this PR does not duplicate planned core work - [x] I have searched GitHub for duplicate or related PRs and linked them above - [x] I have either (a) linked existing issues with `Fixes: #` / `Closes #` / `Refs #` OR (b) described the issue in-PR following the relevant issue template - [x] I have not referenced internal/instance-local Paperclip issues or links (only public GitHub `#NNN` / `github.com/paperclipai/paperclip` URLs) - [x] My branch name describes the change (e.g. `docs/...`, `fix/...`) and contains no internal Paperclip ticket id or instance-derived details - [x] I have run tests locally and they pass - [x] I have added or updated tests where applicable - [x] I have updated relevant documentation to reflect my changes - [x] I have considered and documented any risks above - [ ] All Paperclip CI gates are green - [ ] Greptile is 5/5 with no open P2s, recommendations, or follow-ups - [x] I will address all Greptile and reviewer comments before requesting merge 🤖 Generated with [Claude Code](https://claude.com/claude-code) --------- Co-authored-by: Claude Fable 5 <noreply@anthropic.com> |