Signals: Adds Reference Signals Implementation by jlukic · Pull Request #148 · Semantic-Org/Semantic-Next

jlukic · 2026-04-15T19:55:34Z

Adds safety param to signals, currently defaults to reference

Foundation for three-mode signal safety (freeze/reference/none). defaultSafety starts at 'freeze'; setDefaultSafety validates against the valid set. No consumers yet — the Signal class will wire up in a follow-up.

vercel · 2026-04-15T19:55:42Z

The latest updates on your projects. Learn more about Vercel for GitHub.

Project	Deployment	Actions	Updated (UTC)
semantic-next	Ready	Preview, Comment	May 7, 2026 8:13pm

1 Skipped Deployment

Project	Deployment	Actions	Updated (UTC)
mcp	Ignored	Preview	May 7, 2026 8:13pm

Three-mode safety: 'freeze' (default), 'reference', 'none'. Reads return the protected ref directly; no more per-read cloning. Mutations on frozen values throw at the call site instead of being silently swallowed. - protect() gate: freeze → deepFreeze, otherwise pass-through - value getter/peek() return currentValue directly (no readCopy) - get() loses its options form — use .peek() or .clone() for escapes - new .clone() method returns a deep copy; tracked like .get() - in-place helpers (push/unshift/splice/setIndex/removeIndex/ setArrayProperty/setProperty) branch on safety so reference/none retain O(1) mutation; freeze path rebuilds via spread - mutate(fn): return-value path sets; undefined path notifies (in-place mutation only works under reference/none; throws on freeze) - legacy allowClone: false maps to 'reference' for back-compat Reactivity package tests pass. Downstream callsites that mutated through peek() or the data context will surface as loud failures under freeze; those need per-callsite fixes in follow-up commits.

Replace mutate-in-map with spread-in-map so enrichment helpers don't mutate the caller's input. Two callsites surfaced under freeze-on-set: - mobile-menu addNavIcons: was mutating item.navIcon on items from settings.menu, silently modifying the caller's menu tree - nav-menu addSelectedIndex: was mutating item.selectedIndex on items from settings.menu, same issue; inner .map passes now use the returned value instead of relying on side effects

PR #148 (signal-safety refactor under packages/reactivity) exposed the per-package scoping bug in discover. The prior logic looked up tachometer configs inside the specific changed package and ran their matrix cells — but every tachometer config in this repo lives under packages/renderer, so a PR that only changes packages/reactivity resolved to an empty matrix and the bench workflow skipped silently. This is wrong in the base case: reactivity imports renderer consumes via workspace symlinks, so a signal-path change moves renderer's bench numbers. Same for templating → rendering, utils → everything, etc. The benches are end-to-end measurements that pull the whole framework through a workload; treating them as scoped-by-touched-package was premature. New logic: any trigger that reaches this workflow (push to main, or a pull_request that matched our path filter) runs every tachometer-ci*.json it finds under packages/*/bench/tachometer/. Path filter already restricts triggers to perf-relevant changes, so this doesn't over-fire. The push-branch and pull-request branches collapse to the same one-liner, which is a good sign. Ship.

Build: Discover runs all benchmarkable packages, not just touched ones PR #148 (signal-safety refactor under packages/reactivity) exposed the per-package scoping bug in discover. The prior logic looked up tachometer configs inside the specific changed package and ran their matrix cells — but every tachometer config in this repo lives under packages/renderer, so a PR that only changes packages/reactivity resolved to an empty matrix and the bench workflow skipped silently. This is wrong in the base case: reactivity imports renderer consumes via workspace symlinks, so a signal-path change moves renderer's bench numbers. Same for templating → rendering, utils → everything, etc. The benches are end-to-end measurements that pull the whole framework through a workload; treating them as scoped-by-touched-package was premature. New logic: any trigger that reaches this workflow (push to main, or a pull_request that matched our path filter) runs every tachometer-ci*.json it finds under packages/*/bench/tachometer/. Path filter already restricts triggers to perf-relevant changes, so this doesn't over-fire. The push-branch and pull-request branches collapse to the same one-liner, which is a good sign. Ship.

semantic-performance-bot · 2026-04-15T20:52:56Z

🟡 Mixed (mostly faster) for `e402028` on Benchmark Suite 📊

Base: main · Action: #25519451257 · Raw: bench-report.json

^{Signals: Adds Reference Signals Implementation}

Warning

This PR improves ✅ 19 tests while regressing on ❌ 17 tests.

✅ 19 faster · ❌ 17 slower · 🔍 13 unsure · ⚪ 17 no change

✅ Faster (19) — top 5 shown

Metrics where this PR confidently improved performance compared to main.

metric	Improvement
`signal:reactive-set-index-300`	-99% (114ms) 🏆
`signal:reactive-list-filter-1000x300`	-95% (119ms) 🏆
`signal:reactive-set-property-by-id-200`	-88% (195ms) 🏆
`signal:reactive-push-2000x20`	-83% (190ms) 🏆
`todo:remove-last-100`	-73% (36ms) 🌟

Show all 19 faster metrics

metric	Improvement
`signal:reactive-set-index-300`	-99% (114ms) 🏆
`signal:reactive-list-filter-1000x300`	-95% (119ms) 🏆
`signal:reactive-set-property-by-id-200`	-88% (195ms) 🏆
`signal:reactive-push-2000x20`	-83% (190ms) 🏆
`todo:remove-last-100`	-73% (36ms) 🌟
`todo:rename-500`	-66% (193ms) 🌟
`todo:remove-50-back`	-66% (11ms) 🌟
`todo:toggle-first-100`	-63% (40ms) 🌟
`todo:toggle-middle-100`	-47% (21ms) 🌟
`todo:toggle-last-100`	-45% (26ms) 🌟
`todo:toggle-100`	-44% (19ms) 🌟
`todo:add-20`	-13% (1ms)
`signal:computed-chain-10x60k`	-11% (23ms)
`signal:set-same-10m`	-7% (2ms)
`signal:reactive-multi-read-5x160k`	-6% (13ms)
`hydrate:helper-100-state-change-1k`	-6% (0ms)
`todo:bulk-add-500`	-6% (11ms)
`template:subtemplate-helpers-light-100`	-4% (14ms)
`signal:reaction-coalesce-400x100`	-3% (1ms)

❌ Slower (17) — top 5 shown

Metrics where this PR confidently regressed performance compared to main.

metric	Regression
`todo:filter-cycle-20`	+25% (86ms) ❗
`todo:remove-50-front`	+22% (86ms) ❗
`todo:remove-50-middle`	+20% (39ms) ❗
`todo:edit-cycle-5`	+19% (25ms) ❗
`todo:edit-start-10`	+18% (25ms) ❗

Show all 17 slower metrics

metric	Regression
`todo:filter-cycle-20`	+25% (86ms) ❗
`todo:remove-50-front`	+22% (86ms) ❗
`todo:remove-50-middle`	+20% (39ms) ❗
`todo:edit-cycle-5`	+19% (25ms) ❗
`todo:edit-start-10`	+18% (25ms) ❗
`todo:toggle-all-200`	+16% (338ms) ❗
`krausest:clear-10k`	+15% (16ms) ❗
`todo:remove-first-100`	+9% (128ms)
`template:each-mount-1000`	+8% (3ms)
`signal:reactive-list-replace-1000x1000`	+7% (21ms)
`template:subtemplate-reactive-data-100`	+6% (17ms)
`todo:remove-middle-100`	+6% (47ms)
`todo:clear-completed-250`	+5% (4ms)
`template:subtemplate-data-blob-100`	+5% (12ms)
`template:subtemplate-shorthand-props-100`	+5% (12ms)
`template:snippet-in-subtemplate-100x1k`	+3% (39ms)
`template:subtemplate-helpers-heavy-100`	+3% (14ms)

⚪ No Change (17)

Metrics where this PR measured within ±2% of main — no meaningful performance change detected.

metric	Change
`compiler-micros:ast-walk-15k`	-1.2% – +0.7%
`renderer-micros:build-html-string-10k`	-0.3% – +1.8%
`krausest:create-10k`	-1.3% – -0.3%
`renderer-micros:dom-walker-1000x15`	-1.3% – +1.1%
`hydrate:each-100`	-1.6% – -0.0%
`hydrate:each-100-mount`	-1.3% – +0.4%
`renderer-micros:expr-lisp-50k`	-2.0% – +0.2%
`hydrate:helper-100-mount`	-1.9% – +0.3%
`compiler-micros:parse-cold-complex-200`	-1.6% – +0.5%
`compiler-micros:parse-cold-normal-500`	-0.8% – +1.9%
`signal:reaction-flush-noop-5m`	-0.8% – +1.9%
`signal:reactive-fanout-500x1200`	-0.8% – +1.1%
`krausest:remove-row-front-20`	-0.9% – +1.3%
`krausest:remove-row-middle-20`	-0.8% – +1.9%
`krausest:replace-1k`	-1.3% – +1.0%
`krausest:select-40`	-1.1% – +0.3%
`compiler-micros:snippet-args-5k`	-1.1% – +0.5%

🔍 Unsure (13)

Too Fast to Measure Precisely (13)

On benches this short, OS jitter, GC, and JIT pauses drown out anything under 4%. Bigger changes than that still show up.

metric	Change	Test Time	Expected Noise
`template:active-indicator-200`	-3.2% – +0.6%	~32ms	±5%
`template:active-indicator-nested-200`	-5.0% – -0.9%	~22ms	±5%
`krausest:append-1k`	-6.0% – -1.4%	~171ms	±5%
`krausest:create-1k`	-3.7% – -0.7%	~127ms	±2%
`renderer-micros:expr-js-10k`	-2.9% – -0.5%	~18ms	±6%
`renderer-micros:expr-simple-100k`	-2.7% – -0.7%	~32ms	±5%
`signal:reaction-dep-diff-45k`	-4.5% – -1.4%	~34ms	±4%
`krausest:remove-row-back-100`	-0.1% – +2.4%	~36ms	±2%
`template:snippet-args-per-key-100`	-3.3% – +4.2%	~6ms	±9%
`template:stable-ref-mutate-500`	-7.5% – -0.8%	~13ms	±9%
`signal:sub-unsub-100k`	-4.0% – -0.5%	~22ms	±6%
`krausest:swap-rows-20`	-3.0% – +3.1%	~116ms	±7%
`krausest:update-10th-50`	-2.0% – +0.6%	~150ms	±2%

📖 Bench glossary (66 metrics)

metric	what it tests
`compiler-micros:ast-walk-15k`	Walks a kitchen-sink AST through optimizeAST 15000 times. Merge, hoist, and recurse pass.
`compiler-micros:parse-cold-complex-200`	Compiles a feature-dense kitchen-sink template 200 times. Catches parser regressions on uncommon block paths.
`compiler-micros:parse-cold-normal-500`	Compiles a TodoMVC-style component template 500 times. Headline metric for normal-component compile throughput.
`compiler-micros:snippet-args-5k`	Parses four representative subtemplate-call shapes 5000 times each. Snippet args extraction.
`hydrate:each-100`	Reassigns the items of a hydrated 1000-item list to a fresh array with the same keys and data.
`hydrate:each-100-mount`	Hydrates a server-rendered 1000-item list and waits for it to become interactive without re-rendering.
`hydrate:helper-100-mount`	Hydrates a 1000-item list where each item calls a helper that reads state shared across the list.
`hydrate:helper-100-state-change-1k`	Walks the shared activeID across every item in a hydrated 1000-item list so two items repaint per cycle.
`krausest:append-1k`	Appends 1000 new rows onto an existing 1000-row table.
`krausest:clear-10k`	Clears a 10000-row table back to empty in a single operation.
`krausest:create-10k`	Renders a fresh 10000-row table into an empty parent at ten times the create-1k scale.
`krausest:create-1k`	Renders a fresh 1000-row table into an empty parent.
`krausest:remove-row-back-100`	Removes the last row 100 times from a 1000-row table, with no other rows needing to move.
`krausest:remove-row-front-20`	Removes the first row 20 times from a 1000-row table, with all remaining rows sliding up each time.
`krausest:remove-row-middle-20`	Removes the middle row 20 times from a 1000-row table, with the rows below it sliding up each time.
`krausest:replace-1k`	Replaces 1000 rows with a fresh 1000-row set, diffing the keyed list against a populated table.
`krausest:select-40`	Highlights one row at a time across 40 rows so only the previous and newly highlighted rows update.
`krausest:swap-rows-20`	Swaps the second and second-to-last rows in a 1000-row table, repeated 20 times.
`krausest:update-10th-50`	Updates the label on every tenth row of a 1000-row table, looped 50 times to lift the work above noise.
`renderer-micros:build-html-string-10k`	Builds the HTML string for a realistic card AST 10000 times. Raw assembly throughput.
`renderer-micros:dom-walker-1000x15`	Runs bindMarkers across a 1000-node card fragment 15 times. TreeWalker pass and binding dispatch.
`renderer-micros:expr-js-10k`	Evaluates one arithmetic expression and one ternary 10000 times each. JS-eval hot path.
`renderer-micros:expr-lisp-50k`	Evaluates one Lisp-style helper call 50000 times. Parse-cache lookup and helper dispatch.
`renderer-micros:expr-simple-100k`	Evaluates one simple identifier and one dotted path 100000 times each. Property-lookup hot path.
`signal:computed-chain-10x60k`	Propagates a value change from root to leaf through a 10-deep chain of derived signals 60000 times.
`signal:reaction-coalesce-400x100`	Sets one signal 100 times then flushes once across 400 bursts so 100 subscribers wake one time per burst.
`signal:reaction-dep-diff-45k`	Toggles which of two signals a subscriber reads across 45000 cycles. Per-run dep-set diffing.
`signal:reaction-flush-noop-5m`	Calls Reaction.flush() 5000000 times with no pending work. Scheduler dispatch overhead.
`signal:reactive-fanout-500x1200`	Fans out one signal's value change to 500 subscribers across 1200 successive updates.
`signal:reactive-list-filter-1000x300`	Changes a search-term signal 300 times, re-scanning a 1000-item list on each change.
`signal:reactive-list-replace-1000x1000`	Replaces a 1000-item list signal with a fresh 1000-item array and rescans it 1000 times.
`signal:reactive-multi-read-5x160k`	Changes five signals in turn for 32000 rounds with one subscriber reading all five.
`signal:reactive-push-2000x20`	Appends 20 items onto an empty list signal with a subscriber, across 2000 reset cycles.
`signal:reactive-set-index-300`	Replaces one item by index in a 1000-item list signal across 300 updates, with a subscriber.
`signal:reactive-set-property-by-id-200`	Finds an item by id and updates one field in a 1000-item list signal across 200 alternating updates.
`signal:set-same-10m`	Sets a signal to its current value 10000000 times. Exercises the no-op fast path when nothing changes.
`signal:sub-unsub-100k`	Creates and tears down a subscriber on one signal across 100000 cycles. Subscription churn cost.
`template:active-indicator-200`	Cycles selectedId across 200 list items. Only the previously and newly active items update their class.
`template:active-indicator-nested-200`	Cycles currentUrl through 50 leaf urls in a 5×10×4 nav. Only the previously and newly active leaves should update their…
`template:each-mount-1000`	Mounts a fresh 1000-item each block with five-field items so per-record allocation cost dominates the wall clock.
`template:snippet-args-per-key-100`	Mutates one snippet arg's source across 100 invocations. Adjacent no-signal expressions stay quiet.
`template:snippet-in-subtemplate-100x1k`	Mutates one subtemplate prop's source across 25 cards each invoking 4 inner snippets, 1000 cycles. Snippet bodies shoul…
`template:stable-ref-mutate-500`	Replaces one item by index in a 500-item list across 100 cycles. Only that item's expressions re-render.
`template:subtemplate-data-blob-100`	Mutates one field inside data=expression on 100 children. Every child re-renders by design.
`template:subtemplate-helpers-heavy-100`	100 subtemplates, 4 inner bindings where three call helpers shaped like userland reality — Intl.NumberFormat, Array.fin…
`template:subtemplate-helpers-light-100`	100 subtemplates, 4 inner bindings each calling formatDate / classIf / capitalize. Mutates one source signal — under pe…
`template:subtemplate-reactive-data-100`	Mutates one verbose reactiveData field across 100 child subtemplates. Only the changed field re-evaluates.
`template:subtemplate-shorthand-props-100`	Mutates one shorthand prop's source across 100 child subtemplates. Only that prop re-evaluates.
`todo:add-20`	Appends 20 todo items one at a time, like a user typing entries in a row.
`todo:bulk-add-500`	Renders 500 todo items added at once from a single data load.
`todo:clear-completed-250`	Clears 250 completed items from a 500-item list in one action, like clicking clear completed.
`todo:edit-cycle-5`	Runs 5 full edit-then-save cycles on different items, like editing a row and saving it.
`todo:edit-start-10`	Enters edit mode on 10 different items in a row, like double-clicking each one.
`todo:filter-cycle-20`	Cycles through active, completed, and all filters 20 times on a 100-item list.
`todo:remove-50-back`	Deletes 50 items from the end of a 100-item list, one click at a time.
`todo:remove-50-front`	Deletes 50 items from the front of a 100-item list, one click at a time.
`todo:remove-50-middle`	Deletes 50 items from the middle of a 100-item list, one click at a time.
`todo:remove-first-100`	Deletes the first item 100 times from a 200-item list, with remaining items moving up each time.
`todo:remove-last-100`	Deletes the last item 100 times from a 200-item list, with no other items needing to move.
`todo:remove-middle-100`	Deletes the middle item 100 times from a 200-item list, walking halfway through to find each target.
`todo:rename-500`	Renames items in a 100-item list 500 times via single-field setProperty without editingId co-fires.
`todo:toggle-100`	Cycles through the first 10 items 10 times each, like a user toggling items repeatedly down a list.
`todo:toggle-all-200`	Toggles all 100 items completed and back across 200 cycles via the master checkbox.
`todo:toggle-first-100`	Toggles the first item in a 100-item list 100 times, alternating completed on and off.
`todo:toggle-last-100`	Toggles the last item in a 100-item list 100 times, alternating completed on and off.
`todo:toggle-middle-100`	Toggles a middle item in a 100-item list 100 times, alternating completed on and off.

_{Sample size: 50 floor / 290 max · Noise floor: ±2% · Timeout: 3min · Wall-clock: 19m34s}

Surfaces when a signal-backed data blob flows into the templating boundary under freeze-default: - extend was copying source property descriptors verbatim, so a frozen source produced a fresh target with {writable: false, configurable: false} on every property. Target was extensible but key-level writes threw. Now extend installs a fresh writable descriptor for data properties and only preserves descriptors for accessor (getter/setter) properties. - template.setDataContext thaws this.data lazily when it arrives frozen (via a Template clone with the frozen data argument); assignInPlace then mutates in place as before, preserving reference identity and key-level diffing semantics that the spurious-reactivity guards depend on. - ssr-hydration test updated to clone the array before mutating (the old pattern silently mutated the peek() return).

…tic-Next into perf/signal-safety-v2

Static surface for configuring defaults, discoverable via Signal.* tab completion and validated on assignment: - Signal.equalityFunction / Signal.cloneFunction — accessors with typeof check; throw TypeError on non-function - Signal.safety / Signal.tracing / Signal.stackCapture — accessors delegating to helpers module-level state - Signal.configure({...}) — bulk wrapper; forwards each key through its accessor so validation runs - Signal.defaults — snapshot getter of current settings Drop the allowClone option. The transitional mapping from allowClone: false → safety: 'reference' is removed. Callsites migrated: - lit/renderer.js dataVersion — collapses cleanly to safety: 'none' (allowClone:false + equalityFunction always-false) - native/blocks/each.js itemSignal — safety: 'reference' - templating/src/template.js settingsVars — safety: 'reference' - bench.js / bench-todo.js / tools/benchmark — safety: 'none' Drop legacy function-form statics (Signal.setTracing, Signal.isTracing, Signal.setStackCapture, Signal.isStackCapture). The tracing flag feature was never released; no compat needed.

noop becomes a true no-op — returns undefined, swallows arguments. identity is the new export for the "return first argument unchanged" semantic the old noop was secretly providing. Callsites that depended on the identity behavior migrated to identity: - utils/src/strings.js joinWords transform — the return value feeds the quoting path - component/define-component.js createComponent — return becomes the instance methods (and when no createComponent is provided, the identity default carries callParams into the instance, which the SSR env-guard path depends on) - query/behavior.js createBehavior — same shape as createComponent - query/behavior.js customInvocation — fallback method-not-found handler; return value is the found result All other `= noop` callsites are lifecycle side-effect callbacks (onCreated, onRendered, etc.) whose return values are discarded — true noop is correct there. CHANGELOG entries added for the Reactivity + Utils breaking changes that landed in this release.

…aths - template.js setDataContext: remove Object.isExtensible check. Ran 200-300x per each-block reconcile on subtemplate renders. Local 4-sample A/B measurements show removing it eliminates the remove-middle +11.7% and remove-last +13% regressions identified from aggregated CI artifacts. - signal.js: revert inlined protect() in value setter (speculative, not diagnostic-backed). Keep protect() as a method. - signal.js: revert setArrayProperty single-index fast-path (speculative). - bench-todo.js: scale short benches (toggle/edit/remove single-ops) 10x so they run 120-360ms instead of 6-20ms. Short benches were in tachometer's noise band and rotated between confident-regressed and unsure across runs. With the default safety now 'reference', no signal values are frozen under default use, so template.data will be mutable from the start and setDataContext never needs to detect+replace a frozen target. Users who opt into safety='freeze' can still hit this edge case — if that surfaces, fix it at the signal boundary (don't pass frozen values to Template constructor) rather than defensive checks in a per-call path.

…ignals

…xtend docstring hasProperty becomes a thin re-export of Object.hasOwn (ES2022); proxyObject default switches from noop to an explicit () => ({}); extend gets a docstring capturing its modern-JS shallow-merge semantics.

Shared mutable state now lives on a single config object in helpers.js; Signal's tracing/safety/stackCapture accessors touch it directly instead of routing through setter helpers. Housekeeping statics moved to the bottom of signal.js so the constructor and happy path read at the top, and the mutate() reference/none dedupe path restores its snapshot+equality guard.

Captures decisions for the next utils release window (returnsSelf/returnsTrue/etc. family, remove `any` alias, fix first/last return-type polymorphism). Agent lessons gains "Quiet Code Over Ornamented Code" warning against SCREAMING_CAPS, underscore-prefixes, and dispatcher-layer patterns.

The test enumerated packages/ raw, so stray harness dirs like .claude/ broke the dist/cdn/ existence check.

Bracket assignment already routes through the safety/tracing/stackCapture setters, so the explicit if-ladder was ornament. Agent lessons updated to frame the quiet-code check as a changeset review pass, not just a write-time discipline.

Drops session-scaffolding narration from signal.js, helpers.js, and siblings; keeps load-bearing WHY notes (WeakRef self-stop, captureStackTrace cost, mutate behavior by safety preset). Signal class reorganized with section dividers (Core / Complex / Mutation Helpers / Tracing / Instance of / Configuration), signalTag moved to helpers.js, scheduler's flushTask wrapper replaced by direct Scheduler.flush reference.

Captures the bar for shipped comments (non-obvious to someone who doesn't know the codebase, or a weird trick) with an IS_SIGNAL before/after example, and clarifies that level-2 section dividers for multi-method conceptual clusters are distinct from single-declaration narration labels.

Walk prior bench runs on this PR's branch to build a per-iteration history. The reporter merges it with bench-history.json (main commits) so peak attribution spans BOTH main AND this PR's iterations. An agent iterating on a perf branch now sees: "create-1k peaked at 104.7ms on commit 9bcd3f1 (iteration 10), now at 143.9ms (+37%). Bisect candidates: 442446e, 64c6820, 5595cbd +7 more." New files: - tools/bench-reporter/fetch-pr-history.js — walks gh API for prior successful Benchmarks runs on the PR branch, downloads results-* artifacts, extracts this-change absolute CIs, outputs pr-history.json in bench-history.json schema. Changed files: - reporter.js — new --pr-history flag, mergeHistories() combines main + PR iteration histories sorted by timestamp with SHA dedup. The existing computeHistoryStatus + REOPENED rendering work unchanged. - benchmarks-report.yml — new "Fetch PR iteration history" step before "Generate report" in the comment job. Uses the bot token for API auth. Tested against PR #148's 17 prior runs: 13 REOPENED metrics surfaced with correct peak attribution and bisect candidates.

Refactor: Add safety preset state to reactivity helpers

8004e98

Foundation for three-mode signal safety (freeze/reference/none). defaultSafety starts at 'freeze'; setDefaultSafety validates against the valid set. No consumers yet — the Signal class will wire up in a follow-up.

github-actions Bot added the Reactivity Modifies reactivity package label Apr 15, 2026

vercel Bot deployed to Preview – semantic-next April 15, 2026 19:55 View deployment

github-actions Bot added the Tests Modifies tests label Apr 15, 2026

vercel Bot deployed to Preview – semantic-next April 15, 2026 20:16 View deployment

github-actions Bot added the UI Components Modifies UI components label Apr 15, 2026

vercel Bot deployed to Preview – semantic-next April 15, 2026 20:21 View deployment

jlukic mentioned this pull request Apr 15, 2026

Build: Discover runs all benchmarkable packages (fixes PR #148) #149

Merged

github-actions Bot added the CI modifies continuous integration label Apr 15, 2026

vercel Bot deployed to Preview – semantic-next April 15, 2026 20:43 View deployment

Merge branch 'main' into perf/signal-safety-v2

ecf22c3

github-actions Bot removed the CI modifies continuous integration label Apr 15, 2026

vercel Bot deployed to Preview – semantic-next April 15, 2026 20:55 View deployment

jlukic added 2 commits April 15, 2026 17:13

Merge branch 'perf/signal-safety-v2' of github.com:Semantic-Org/Seman…

27f68db

…tic-Next into perf/signal-safety-v2

github-actions Bot added Templating Modifies templating package Utils Modifies utilities package labels Apr 15, 2026

vercel Bot deployed to Preview – semantic-next April 15, 2026 21:16 View deployment

jlukic added 2 commits April 15, 2026 18:19

github-actions Bot added Query Modifies query package Component Component Rendering Docs Modifies documentation labels Apr 15, 2026

Perf: Inline protect in hot value setter to avoid method call

28901c4

vercel Bot deployed to Preview – semantic-next April 16, 2026 04:30 View deployment

vercel Bot deployed to Preview – semantic-next April 16, 2026 13:14 View deployment

jlukic added 7 commits April 16, 2026 10:19

Refactor: Drop redundant safety:'reference' from framework-internal S…

7ce4e64

…ignals

Docs: Update utils-noop example for post-split semantics

106e66e

Test: Filter dotfile dirs in CDN packages discovery

5ba66ea

The test enumerated packages/ raw, so stray harness dirs like .claude/ broke the dist/cdn/ existence check.

vercel Bot deployed to Preview – semantic-next April 16, 2026 15:31 View deployment

jlukic added 5 commits April 16, 2026 11:59

Perf: remove unnecessary clone on menu

67e529f

Harness: Update perf skill to include rebuttals to perf tests

f197f27

Harness: workspace

6da9d1c

vercel Bot deployed to Preview – semantic-next April 16, 2026 17:11 View deployment

Merge remote-tracking branch 'origin' into perf/signal-safety-v2

64e90a1

vercel Bot deployed to Preview – semantic-next April 18, 2026 12:17 View deployment

jlukic changed the title ~~Refactor: Add safety preset state to reactivity helpers~~ Signals: Adds Reference Signals Implementation Apr 18, 2026

jlukic mentioned this pull request Apr 18, 2026

Build: PR-iteration history for cross-commit peak attribution #152

Merged

Merge branch 'main' into perf/signal-safety-v2

0b4c2e4

vercel Bot deployed to Preview – semantic-next April 18, 2026 12:56 View deployment

Merge branch 'main' into perf/signal-safety-v2

ae052eb

vercel Bot deployed to Preview – semantic-next May 7, 2026 18:12 View deployment

Merge branch 'main' into perf/signal-safety-v2

e402028

vercel Bot deployed to Preview – semantic-next May 7, 2026 20:13 View deployment

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Signals: Adds Reference Signals Implementation#148

Signals: Adds Reference Signals Implementation#148
jlukic wants to merge 38 commits intomainfrom
perf/signal-safety-v2

jlukic commented Apr 15, 2026 •

edited

Loading

Uh oh!

vercel Bot commented Apr 15, 2026 •

edited

Loading

Uh oh!

semantic-performance-bot Bot commented Apr 15, 2026 •

edited

Loading

Too Fast to Measure Precisely (13)

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

1 participant

Conversation

jlukic commented Apr 15, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

vercel Bot commented Apr 15, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

semantic-performance-bot Bot commented Apr 15, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

🟡 Mixed (mostly faster) for e402028 on Benchmark Suite 📊

✅ Faster (19) — top 5 shown

❌ Slower (17) — top 5 shown

Too Fast to Measure Precisely (13)

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

1 participant

jlukic commented Apr 15, 2026 •

edited

Loading

vercel Bot commented Apr 15, 2026 •

edited

Loading

semantic-performance-bot Bot commented Apr 15, 2026 •

edited

Loading

🟡 Mixed (mostly faster) for `e402028` on Benchmark Suite 📊