* Add selection-driven span-detail primitives
Extract the run/step/hook/sleep fetch+hydrate core out of useWorkflowResourceData
into a plain async fetchSpanDetailResource (no React state), and add a
selection-driven state machine in web-shared:
- deriveSpanDetailView / resourceNeedsFetchedDetail: pure view-model deriver
whose status (idle/loading/ready/error) is a function of (selection, fetched
detail), so it can never lag the selection.
- useSelectedSpanDetail: fetches a selected span's detail directly with a
request-token to drop stale/out-of-order responses.
* Drive trace detail panel from the span-detail state machine
Replace the cross-package selection round-trip (EntityDetailPanel useEffect ->
onSpanSelect -> page spanSelection state -> useWorkflowResourceData -> context)
with a single injected fetchSpanDetail capability:
- EntityDetailPanel consumes useSelectedSpanDetail; its loading state now stays
in phase with the selected span, so Input/Output no longer vanish and pop back
in while navigating.
- SidebarDataContext drops spanDetailData/Loading/Error + onSpanSelect for a
single fetchSpanDetail; RunDetailView injects it and drops the duplicate
spanSelection state.
- WorkflowTraceViewer / RunTraceView take fetchSpanDetail too.
* Test span-detail view-model transitions; add changeset
Cover deriveSpanDetailView (idle/loading/ready/error, stale-detail rejection,
hooks ready inline) and resourceNeedsFetchedDetail.
* Trim redundant/narrative comments in span-detail state machine
Comment-only cleanup: drop PR-narration and cross-file duplication from the
deriveSpanDetailView / useSelectedSpanDetail / EntityDetailPanel / fetchSpanDetail
doc comments, keeping the non-obvious intent (request-token, error scoping,
decrypt closure).
* delete pointless coments
---------
Co-authored-by: Cursor Agent <cursoragent@cursor.com>
- Add deprecation banners (with migration-guide link) to the DurableAgent and
WorkflowChatTransport API references in v4 and v5; keep the full API surface intact
- Bring v4 headline guides to parity with v5's WorkflowAgent migration (ai/index,
foundations/streaming, the cookbook recipe + index)
- Convert standard agent examples (defining-tools, message-queueing) to WorkflowAgent
and reframe the streamText-vs comparison page
- Banner + repoint the deep recipes that stream custom UIMessageChunk data parts
(chat-session-modeling, human-in-the-loop, agent-cancellation, serializable-steps) —
that pattern doesn't map to WorkflowAgent's ModelCallStreamPart model, so their
legacy DurableAgent examples are kept behind a clear deprecation banner
- Point all WorkflowChatTransport examples at the @ai-sdk/workflow 1:1 port
- Rename the cookbook agent-patterns recipe to WorkflowAgent
Co-authored-by: Claude Opus 4.8 <noreply@anthropic.com>
Move cn to its own lib/cn module that teaches tailwind-merge about the
design-system text-heading/text-label/text-copy/text-button/material
utilities, and import it directly from lib/cn.
Co-authored-by: Cursor Agent <cursoragent@cursor.com>
* Propagate trace context to vercel-workflow.com in workbench instrumentation
@vercel/otel only propagates W3C trace context to Vercel deployment URLs
by default, so outgoing requests to the workflow-server
(vercel-workflow.com) got a client span with no `traceparent` header —
breaking the APM trace link to workflow-server's spans. Add
`instrumentationConfig.fetch.propagateContextUrls` for the workflow-server
domain in every workbench that uses @vercel/otel: example,
nextjs-turbopack, nextjs-webpack, and sveltekit. The Next.js and SvelteKit
apps already declared @vercel/otel but weren't registering it at all; they
now do.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
* Also propagate trace context to the Vercel Queue Service (vercel-queue.com)
The workflow-server queue path (@vercel/queue) sends to regional
vercel-queue.com subdomains (e.g. iad1.vercel-queue.com) when not using the
queues proxy, which were missing a `traceparent` header for the same reason
as vercel-workflow.com. Add `/vercel-queue\.com/` to propagateContextUrls in
all four workbench instrumentation configs.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 4.8 <noreply@anthropic.com>
* Use text-gray-900 for trace shortcut helper text
Co-authored-by: Mitul Shah <mitulxshah@gmail.com>
* Add changeset for trace shortcut helper text color
Co-authored-by: Mitul Shah <mitulxshah@gmail.com>
---------
Co-authored-by: Cursor Agent <cursoragent@cursor.com>
* fix(web-shared): stop data inspector duplicating expanded objects
Expanded objects/arrays now render bracket delimiters ({ … } / [ … ])
instead of repeating the inline preview alongside the child tree.
Co-authored-by: Cursor <cursoragent@cursor.com>
* nice
* fix: sync lockfile after dropping react-inspector
The react-inspector removal landed in package.json but the lockfile was
reverted during cleanup, breaking frozen-lockfile installs in CI.
Co-authored-by: Cursor <cursoragent@cursor.com>
* fix(web-shared): keep data inspector colors theme-aware
Drop the dark-mode color overrides (and the data-theme/useDarkMode
wiring) and rely on the theme-aware --ds-* tokens, matching front:
strings stay green in both light and dark instead of turning blue.
Co-authored-by: Cursor <cursoragent@cursor.com>
* test(web-shared): add data inspector tests; tidy comments
Cover collapseRefs ref/typed-array/Map/Set handling and the rendered
tree (keys, value colors, brackets, commas, collapse/expand, empties,
dates, class/Map/Set prefixes) via jsdom + testing-library.
Co-authored-by: Cursor <cursoragent@cursor.com>
* revert(web-shared): drop data inspector tests
Remove the test suite and its jsdom/@testing-library devDependencies to
avoid adding new packages. Keeps the data inspector code unchanged.
Co-authored-by: Cursor <cursoragent@cursor.com>
* refactor(web-shared): extract data inspector styles to a sibling module
Move the class-name map and CSS string out of the component into
data-inspector.styles.ts for readability. Still injected via the
hoistable <style>; no behavior or dependency change.
Co-authored-by: Cursor <cursoragent@cursor.com>
* fix(web-shared): restore ARIA tree semantics for the data inspector
Co-authored-by: Cursor <cursoragent@cursor.com>
* fix(web-shared): render RegExp values as /source/flags
Co-authored-by: Cursor <cursoragent@cursor.com>
* test(web-shared): add unit tests for CopyableDataBlock JSON viewer (#2584)
Export serializeForClipboard and cover its clipboard formatting (strings, primitives, pretty-printed JSON, circular/BigInt fallbacks) plus CopyableDataBlock/EncryptedDataBlock rendering.
Co-authored-by: Cursor Agent <cursoragent@cursor.com>
---------
Co-authored-by: Cursor <cursoragent@cursor.com>
Swap the text-based eve placeholder for the real eve wordmark (hard-copied
SVG from @vercel/geistcn-assets, themed via currentColor) and drop Streamdown
so AI Elements is last.
Co-authored-by: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Co-authored-by: Pranay Prakash <pranay.gp@gmail.com>
* Add Platformatic World to worlds-manifest.json
Signed-off-by: marcopiraccini <marco.piraccini@gmail.com>
* ci fixup
Signed-off-by: marcopiraccini <marco.piraccini@gmail.com>
* ci: pin platformatic world image to 0.8.1 and harden community-world runner
Signed-off-by: marcopiraccini <marco.piraccini@gmail.com>
* platforamtic-world version
Signed-off-by: marcopiraccini <marco.piraccini@gmail.com>
* ci: wire generic docker service-type into community benchmark workflow
The shared community-worlds matrix now emits service-type "docker" for any
world with non-builtin or multiple services (e.g. Platformatic, which needs
postgres + the platformatic/workflow image). tests.yml's e2e-community path
already handles it, but benchmarks.yml's benchmark-community path
(label-gated, non-blocking) did not — so a "community-benchmarks" run would
start no services and fail.
Mirror the e2e "Start Docker services" step, package-version pin, and docker
cleanup into benchmark-community-world.yml, and pass `services`/`version`
through from benchmarks.yml.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
---------
Signed-off-by: marcopiraccini <marco.piraccini@gmail.com>
Co-authored-by: Pranay Prakash <pranay.gp@gmail.com>
Co-authored-by: Claude Opus 4.8 <noreply@anthropic.com>
Section index card grids (e.g. foundations) were hand-written and drifted
from the sidebar (meta.json) and the actual pages. Make them derive from
the fumadocs page tree (single source of truth) and add CI lint so the
card grid and navigation can't fall out of sync again.
- resolveSectionChildren + <AutoCards/>, bound in both v4 and v5 docs
routes (correct /docs vs /v5/docs URL spaces)
- getLLMText expands <AutoCards/> so llms.txt/.md/copy-page keep child links
- manualCards frontmatter opt-out for curated pages (source.config.ts)
- checkSectionCards (card<->nav completeness) + checkMetaEntriesResolve
(dangling meta entries) in scripts/lint.ts
- convert foundations + errors (drift fixes) and v5 observability to AutoCards
- mark deploying + ai as manualCards (intentionally curated)
- remove dangling meta entries: v4 cancellation (x2), root introduction
(x2), v4/internal serializable-abort-controller
Co-authored-by: Claude Opus 4.8 <noreply@anthropic.com>
* perf(core): memoize step return value hydration across replays
The inline replay loop re-executes the workflow body and re-consumes the
full event log on every iteration. For each already-completed step, the
step consumer re-decrypted and re-devalue-parsed the serialized result on
every replay — O(N^2) decrypt+parse operations across a single
invocation of a sequential N-step workflow.
Add a per-run memoization cache, owned by the inline loop in runtime.ts
(alongside cachedEvents) so it survives across replay iterations of the
same run but never leaks across runs. It is threaded into runWorkflow and
stored on the orchestrator context, and consulted in the step_completed
path keyed by the persisted event id. This makes a completed step's
hydrated result O(1) on subsequent replays, turning the aggregate cost
into O(N).
Determinism is preserved: the cache lookup happens inside the existing
ctx.promiseQueue slot and still resolves via the same resolve(), so a
cache hit occupies the identical position in the ordered delivery chain a
re-hydrate would have — pendingDeliveries accounting, delivery barriers,
and Promise.race/all replay are untouched.
Identity safety: hydrateStepReturnValue returns a fresh object graph each
call and each replay runs in a fresh VM, so sharing an object reference
across replays could let one replay's mutation leak into the next. Only
primitive results are memoized (immutable, reference-share == re-parse);
non-primitives re-hydrate fresh every replay, exactly as before. Hook,
wait, and abort hydration paths are intentionally left uncached.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
* perf(core): bound memoized step-hydration cache by primitive size
Address the review note that the per-run step hydration cache was never
size-bounded: cached entries hold the decrypted/parsed plaintext of a
primitive step result for the whole invocation, on top of the serialized
bytes already retained in cachedEvents, so a long run returning large
strings could roughly double peak retained memory for those results.
Document the cache's memory characteristic (per-invocation, freed when the
invocation ends, bounded by primitive-returning step count) and cap the
only primitive types that can carry a large payload: string/bigint results
longer than MAX_MEMOIZED_PRIMITIVE_LENGTH (4 KiB) fall through to the
existing per-replay re-hydrate path instead of being memoized. Large
payloads are cheap to re-hydrate relative to their footprint, so this caps
the worst case at negligible cost. Other primitives are inherently small
and always memoized.
The cap only ever reduces what is cached, so deterministic replay is
unaffected: oversized values take the already-correct re-hydrate path.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 4.8 <noreply@anthropic.com>
github-actions[bot]