The Phase 3 virtualizer (PAP-2645) used useWindowVirtualizer, but on the
real issue page the chat lives inside `<main id="main-content"
overflow-auto>`. Window scroll never moves on desktop, so the virtualizer
only ever mounted the first 7 rows and the chat area went blank past the
first viewport. Detect the actual scroll container at mount via DOM walk
and rekey the inner virtualizer to use element-mode observers/scroll-fn
when an overflow ancestor exists; fall back to window mode for mobile and
the auth-free perf fixture, where the document scrolls.
Adds a regression test that mounts IssueChatThread inside an overflow-auto
ancestor and asserts jump-to-latest drives that element's scrollTo, not
window.scrollTo.
Co-Authored-By: Paperclip <noreply@paperclip.ing>
If the issue reply composer would post without any assignee, the first
Send press now surfaces a warning toast instead of submitting. A second
press confirms the user wants to proceed unassigned, and changing the
assignee resets the confirmation.
Co-Authored-By: Paperclip <noreply@paperclip.ing>
Add @tanstack/react-virtual and gate IssueChatThread on a 150-row
threshold. Thread shorter than the threshold still uses the direct
render path; longer threads window the merged feed via
useWindowVirtualizer with dynamic measurement and overscan, so the
document scroll continues to represent the full conversation height
without introducing a nested scrollbar.
Co-Authored-By: Paperclip <noreply@paperclip.ing>
UX feedback split the badge by role: on the source issue the linked pill
describes a state ("Under review"), while on the productivity-review issue
itself the static pill identifies the kind of issue ("Productivity review").
Tooltip and aria-label still surface the formal name on hover.
Co-Authored-By: Paperclip <noreply@paperclip.ing>
Adds a productivityReview lookup to single-issue and list responses, plus
a ProductivityReviewBadge component that surfaces the open productivity
review on the source issue's header (with trigger/no-comment-streak
detail in the tooltip), and a compact eye glyph next to the StatusIcon
on IssueRow so operators can spot yellow tasks in any list. The review
issue itself shows a static "Productivity review" pill in the header
when its originKind matches.
Also extends the shared IssueOriginKind union with the recovery-side
kinds (harness_liveness_escalation, issue_productivity_review,
stranded_issue_recovery) so UI checks against issue.originKind type-check.
Wires the new helper through GET /issues/:id, GET /companies/:id/issues
(via issueService.list), and GET /issues/:id/heartbeat-context. Adds a
status-language Storybook story matrix for UX review.
Co-Authored-By: Paperclip <noreply@paperclip.ing>
Reviewer follow-up — pills sat slightly low after the baseline-to-middle
switch. position: relative + top: -1px raises them to optically center
with the surrounding cap height.
Inline-flex baseline behavior is browser-dependent, which made skill/agent/issue
chips in MarkdownBody and the MDXEditor hang below the surrounding text. Switch
to vertical-align: middle and drop the vertical padding so the pill sits
centered on the line with the prose around it.
Adds a skill-chip line to the MarkdownBody storybook fixture so this regresses
loudly next time.
Co-Authored-By: Paperclip <noreply@paperclip.ing>
Wrap the variables editor in a single rounded container so the
"Variables" header and the per-variable cards share one border with
internal dividers, instead of stacking as two visually separate
containers when expanded.
Add a help button to the variable instructions hint that opens a modal
explaining custom {{variable_name}} placeholders and the built-in
{{date}} and {{timestamp}} variables.
Co-Authored-By: Paperclip <noreply@paperclip.ing>
Renders as a human-readable date and time in UTC (e.g. "April 28,
2026 at 12:17 PM UTC"), complementing the existing {{date}} variable.
Co-Authored-By: Paperclip <noreply@paperclip.ing>
## Thinking Path
> - Paperclip's board UI relies on compact selectors for frequent issue
and agent edits.
> - Inline selectors often live inside larger keyboard-aware surfaces
such as composers and popovers.
> - Arrow, enter, tab, and escape keys handled by the selector should
not leak to parent document shortcuts.
> - Stale company selection should also stay hidden until the company
list confirms it is valid.
> - This pull request tightens inline selector keyboard handling and
adds regression coverage for stale company bootstrap behavior.
> - The benefit is fewer accidental parent interactions and safer
company-scoped UI initialization.
## What Changed
- Added a stable empty `recentOptionIds` default so selector filtering
does not get a new array every render.
- Mirrored highlighted option state into a ref so Enter/Tab commits the
current highlighted option reliably after keyboard navigation.
- Stopped propagation for selector-owned navigation/commit/escape keys.
- Added jsdom regressions for inline selector keyboard handling and
CompanyProvider stale selection behavior.
## Verification
- `pnpm exec vitest run ui/src/components/InlineEntitySelector.test.tsx
ui/src/context/CompanyContext.test.tsx`
- Targeted selector and CompanyProvider tests pass cleanly without React
`act(...)` warnings.
- Screenshots not attached: this is keyboard/state behavior covered by
component tests.
## Risks
- Low risk: changes are scoped to inline selector key handling and
tests. The main behavior shift is intentionally preventing handled
selector keys from reaching parent listeners.
> For core feature work, check [`ROADMAP.md`](ROADMAP.md) first and
discuss it in `#dev` before opening the PR. Feature PRs that overlap
with planned core work may need to be redirected — check the roadmap
first. See `CONTRIBUTING.md`.
## Model Used
- OpenAI Codex coding agent based on GPT-5, tool-enabled local
repository and shell access, Paperclip heartbeat context.
## 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 run tests locally and they pass
- [x] I have added or updated tests where applicable
- [ ] If this change affects the UI, I have included before/after
screenshots
- [x] I have updated relevant documentation to reflect my changes
- [x] I have considered and documented any risks above
- [x] I will address all Greptile and reviewer comments before
requesting merge
---------
Co-authored-by: Paperclip <noreply@paperclip.ing>
## Thinking Path
> - Paperclip operators need fast control over the agents running their
company.
> - The sidebar is the persistent place operators scan agent state while
navigating the board UI.
> - Agent pause and resume already exist as control-plane actions, but
sidebar users had to navigate away to use them.
> - This pull request adds a compact per-agent action menu in the
sidebar.
> - It keeps edit, pause, and resume close to the visible agent list
while preserving existing navigation behavior.
> - The benefit is faster operator intervention when an agent needs to
be paused or restarted.
## What Changed
- Refactored sidebar agent rows into a small item component with a
hover/focus action menu.
- Added edit, pause, and resume actions using existing agent API calls
and cache invalidation keys.
- Added success/error toasts for pause and resume mutations.
- Tracked pause/resume pending state per agent so one active mutation
does not disable every sidebar row.
- Disabled direct sidebar resume for budget-paused agents and labeled
that state clearly.
- Added jsdom coverage for active-agent pause, paused-agent resume,
per-agent pending state, and budget-paused resume protection.
- Added visual review artifacts for the default row and opened action
menu:
- [Sidebar row](docs/pr-screenshots/pr-4616/sidebar-agent-row.png)
- [Sidebar action
menu](docs/pr-screenshots/pr-4616/sidebar-agent-actions.png)
## Verification
- `pnpm exec vitest run ui/src/components/SidebarAgents.test.tsx`
- `pnpm --filter @paperclipai/server prepare:ui-dist`
- Browser screenshot pass against a temporary local trusted instance at
`http://127.0.0.1:3102` using Playwright.
## Risks
- Low risk: UI-only addition using existing agent pause/resume
endpoints. The main risk is layout crowding for very narrow sidebars,
mitigated by the icon-only trigger and existing truncation.
- Budget-paused agents now require a non-sidebar path to resume, which
is intentional to avoid accidentally restarting agents stopped by budget
enforcement.
> For core feature work, check [`ROADMAP.md`](ROADMAP.md) first and
discuss it in `#dev` before opening the PR. Feature PRs that overlap
with planned core work may need to be redirected — check the roadmap
first. See `CONTRIBUTING.md`.
## Model Used
- OpenAI Codex coding agent based on GPT-5, tool-enabled local
repository and shell/browser access, Paperclip heartbeat context.
## 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 run tests locally and they pass
- [x] I have added or updated tests where applicable
- [x] If this change affects the UI, I have included before/after
screenshots
- [x] I have updated relevant documentation to reflect my changes
- [x] I have considered and documented any risks above
- [x] I will address all Greptile and reviewer comments before
requesting merge
---------
Co-authored-by: Paperclip <noreply@paperclip.ing>
## Thinking Path
> - Paperclip coordinates recurring agent work through scheduled and
manual routines.
> - Manual routine runs are board-initiated work and should stay visible
to the human who kicked them off.
> - Routine execution issues are agent-assigned, so they can be filtered
away from a board user's inbox unless the user is recorded as touching
the work.
> - Coalesced or skipped active routine runs have the same visibility
problem because they reuse an existing live issue.
> - This pull request carries the manual runner actor into routine
dispatch and touches the linked issue for that user's inbox.
> - The benefit is that manually triggered routine work stays
discoverable by the operator who started it.
## What Changed
- Passed the board or agent actor from the routine run route into the
routine service.
- Recorded manual board runners as `createdByUserId` on fresh routine
execution issues.
- Touched coalesced or skipped active routine issues for the manual
runner by updating read state and clearing that user's inbox archive.
- Added route and service regressions for manual routine run actor
propagation and inbox visibility.
## Verification
- `pnpm exec vitest run server/src/__tests__/routines-routes.test.ts
server/src/__tests__/routines-service.test.ts`
## Risks
- Low risk: the change is scoped to manual routine runs and only updates
issue attribution/read-state metadata for the initiating actor.
- No migrations.
> For core feature work, check [`ROADMAP.md`](ROADMAP.md) first and
discuss it in `#dev` before opening the PR. Feature PRs that overlap
with planned core work may need to be redirected — check the roadmap
first. See `CONTRIBUTING.md`.
## Model Used
- OpenAI Codex coding agent based on GPT-5, tool-enabled local
repository and shell access, Paperclip heartbeat context.
## 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 run tests locally and they pass
- [x] I have added or updated tests where applicable
- [x] If this change affects the UI, I have included before/after
screenshots
- [x] I have updated relevant documentation to reflect my changes
- [x] I have considered and documented any risks above
- [x] I will address all Greptile and reviewer comments before
requesting merge
---------
Co-authored-by: Paperclip <noreply@paperclip.ing>
## Thinking Path
> - Paperclip orchestrates AI agents for autonomous companies.
> - Agent assignments must reliably turn into heartbeat work without
board operators manually nudging stuck tasks.
> - The stranded-assignment recovery sweep already handles failed or
lost runs.
> - But assigned `todo` issues with no prior run could sit idle because
there was nothing to retry or recover.
> - This pull request dispatches those never-started assigned todos as
normal assignment wakes.
> - The benefit is that recovery fixes missed initial dispatches without
creating unnecessary recovery issues.
## What Changed
- Added an initial assigned-todo dispatch path to the recovery service
when an assigned `todo` issue has no heartbeat run yet.
- Reused invocation budget hard-stop checks before dispatching or
requeueing recovery work.
- Counted `assignmentDispatched` in startup/scheduled recovery logs.
- Added heartbeat recovery regressions for first dispatch, duplicate
queued wake prevention, budget-blocked skips, and paused-agent skips.
## Verification
- `pnpm exec vitest run
server/src/__tests__/heartbeat-process-recovery.test.ts`
## Risks
- Low to medium risk: this changes liveness recovery behavior for
assigned `todo` issues, but it stays on the existing assignment wake
path and skips paused or budget-blocked agents.
- No migrations.
> For core feature work, check [`ROADMAP.md`](ROADMAP.md) first and
discuss it in `#dev` before opening the PR. Feature PRs that overlap
with planned core work may need to be redirected — check the roadmap
first. See `CONTRIBUTING.md`.
## Model Used
- OpenAI Codex coding agent based on GPT-5, tool-enabled local
repository and shell access, Paperclip heartbeat context.
## 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 run tests locally and they pass
- [x] I have added or updated tests where applicable
- [x] If this change affects the UI, I have included before/after
screenshots
- [x] I have updated relevant documentation to reflect my changes
- [x] I have considered and documented any risks above
- [x] I will address all Greptile and reviewer comments before
requesting merge
## Thinking Path
> - Paperclip orchestrates AI agents for zero-human companies
> - The control plane must recover stranded agent work without creating
new operational loops
> - Stranded recovery issues can themselves fail, and exposing raw retry
errors in comments can leak sensitive adapter details
> - New local companies also should not force a hire-approval gate
unless operators enable that policy
> - This pull request hardens recovery issue handling, redacts retry
failure details in issue copy, preserves `maxConcurrentRuns: 1`, and
flips new-hire approval to an opt-in default
> - The benefit is safer automatic recovery and smoother default company
setup without hidden migration conflicts
## What Changed
- Added migration `0071_default_hire_approval_off` and updated company
schema/import/export/docs so hire approvals default off and serialize
only when enabled.
- Added migration `0072_large_sandman` with a partial unique index
preventing duplicate active stranded recovery issues for the same source
issue.
- Blocked failed `stranded_issue_recovery` issues in place instead of
creating nested recovery issues.
- Redacted latest retry failure details from recovery issue comments
while still linking reviewers to run evidence.
- Allowed `maxConcurrentRuns: 1` to be honored by heartbeat concurrency
normalization.
- Added focused regression coverage for recovery recursion, redaction,
migration ordering, and concurrency behavior.
## Verification
- `pnpm --filter @paperclipai/db run check:migrations`
- `pnpm exec vitest run --project @paperclipai/server
server/src/__tests__/recovery-classifiers.test.ts`
- `pnpm exec vitest run --project @paperclipai/server
server/src/__tests__/company-portability.test.ts --pool=forks
--poolOptions.forks.isolate=true`
- `pnpm exec vitest run --project @paperclipai/server
server/src/__tests__/agent-permissions-routes.test.ts --pool=forks
--poolOptions.forks.isolate=true`
- `pnpm --filter @paperclipai/server typecheck`
- `pnpm exec vitest run --project @paperclipai/server
server/src/__tests__/heartbeat-process-recovery.test.ts --pool=forks
--poolOptions.forks.isolate=true` exits 0, but this host skipped the
embedded Postgres tests with the existing init guard.
- `pnpm exec vitest run --project @paperclipai/server
server/src/__tests__/heartbeat-dependency-scheduling.test.ts
--pool=forks --poolOptions.forks.isolate=true` exits 0, but this host
skipped the embedded Postgres tests with the existing init guard.
## Risks
- Migration risk is low but this PR intentionally owns both new
migrations to avoid separate PR migration-journal conflicts.
- Recovery comments now require operators to inspect linked run evidence
for details instead of reading raw errors inline.
- The hire approval default changes behavior for newly created/imported
companies only; existing persisted company settings are not changed
except by the SQL default for future rows.
> For core feature work, check [`ROADMAP.md`](ROADMAP.md) first and
discuss it in `#dev` before opening the PR. Feature PRs that overlap
with planned core work may need to be redirected — check the roadmap
first. See `CONTRIBUTING.md`.
## Model Used
- OpenAI Codex, GPT-5 coding agent, tool-enabled terminal/GitHub
workflow, reasoning mode active. Context window not exposed in this
environment.
## 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 run tests locally and they pass
- [x] I have added or updated tests where applicable
- [x] If this change affects the UI, I have included before/after
screenshots
- [x] I have updated relevant documentation to reflect my changes
- [x] I have considered and documented any risks above
- [x] I will address all Greptile and reviewer comments before
requesting merge
---------
Co-authored-by: Paperclip <noreply@paperclip.ing>
## Thinking Path
> - Paperclip orchestrates AI agents for zero-human companies
> - Company skills are part of the reusable agent capability layer
> - Skill inventory refresh work can outlive the company it was
requested for
> - Without an explicit company existence check, stale refreshes can
continue into bundled/local skill cleanup for deleted or missing
companies
> - This pull request makes company-skill listing fail fast when the
company no longer exists
> - The benefit is clearer API behavior and less stale background work
against missing company scope
## What Changed
- Added a company existence check before `companySkillService.list()`
refreshes bundled and local-path skill state.
- Added regression coverage asserting missing companies return `404
Company not found`.
## Verification
- `pnpm exec vitest run --project @paperclipai/server
server/src/__tests__/company-skills-service.test.ts --pool=forks
--poolOptions.forks.isolate=true` exits 0, but this host skipped the
embedded Postgres tests with the existing init guard.
## Risks
- Low risk. Existing callers for valid companies are unchanged.
- Missing-company callers now receive an explicit 404 instead of
continuing refresh work.
> For core feature work, check [`ROADMAP.md`](ROADMAP.md) first and
discuss it in `#dev` before opening the PR. Feature PRs that overlap
with planned core work may need to be redirected — check the roadmap
first. See `CONTRIBUTING.md`.
## Model Used
- OpenAI Codex, GPT-5 coding agent, tool-enabled terminal/GitHub
workflow, reasoning mode active. Context window not exposed in this
environment.
## 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 run tests locally and they pass
- [x] I have added or updated tests where applicable
- [x] If this change affects the UI, I have included before/after
screenshots
- [x] I have updated relevant documentation to reflect my changes
- [x] I have considered and documented any risks above
- [x] I will address all Greptile and reviewer comments before
requesting merge
