From da1e83515055f7647c5034e35d527d810816349c Mon Sep 17 00:00:00 2001 From: Claude Date: Mon, 24 Aug 2026 02:13:58 +0000 Subject: [PATCH 1/3] =?UTF-8?q?prompt:=20scope=20the=20test-performance=20?= =?UTF-8?q?board=20=E2=80=94=20run=20times,=20hangs,=20no=5Frun,=20one-tap?= =?UTF-8?q?=20fix=20prompts?= MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Deep-research scoping for a standing dashboard of the testing/integration infrastructure's run times (PR smoke gates, unit tests, import time), with history to flag drops, kill-timer/hang events, and the NO_RUN census β€” every row carrying its own one-tap Claude prompt. - docs/pyautoheart/test_performance_board_assessment.md β€” the assessment: Heart owns measurement and publishes a `performance` block in board.json (schema v2, additive, rows carry their own prompts); the Brain board consumes the headline verbatim via the existing heart_blockers contract. Three data planes: Actions-API wall-clock scrape (proven by morning_health.yml), per-script timings via the delegated runner (one PyAutoHands change since #260-#263), and Heart's tracked baselines (fix script_timing first β€” its bug prompt is filed and unissued). - draft/feature/pyautoheart/test_performance_board.md β€” the actionable phased prompt. - dashboard regenerated (intake --apply dashboard). Co-Authored-By: Claude Fable 5 Claude-Session: https://claude.ai/code/session_01EoDPz2LevKeBaDwqFKtZrU --- dashboard.html | 25 +- dashboard.md | 42 +-- .../test_performance_board_assessment.md | 281 ++++++++++++++++++ .../pyautoheart/test_performance_board.md | 91 ++++++ 4 files changed, 410 insertions(+), 29 deletions(-) create mode 100644 docs/pyautoheart/test_performance_board_assessment.md create mode 100644 draft/feature/pyautoheart/test_performance_board.md diff --git a/dashboard.html b/dashboard.html index 1bf29b17..9070c653 100644 --- a/dashboard.html +++ b/dashboard.html @@ -123,10 +123,10 @@
πŸ“‹

PyAutoMindDashboard

Intent. Priority. Flow.

Every task the Mind is holding. Tap a task's πŸ“‹ and its /start_dev command is on your clipboard β€” paste it into a Claude Code chat to route Claude straight to that task. Recent is the same work by date β€” what has been happening rather than what to do next.

- +

markdown version

Start here

-

Highest priority (filed as high) β€” showing 12 of 17

+

Highest priority (filed as high) β€” showing 12 of 18

TRIAGE: needs manual review before routing❓ triagemediumsafehigh

Numba CPU likelihood phase 1: batched MGE convolution + operated-matrix caching✨ featureautoarraymediumsupervisedhigh

Rectangular mesh split: Bilinear (fast CPU default) vs RTU (advanced/GPU)✨ featureautoarraymediumsupervisedhigh

@@ -134,11 +134,11 @@

Highest priority (filed as high) β€” showing 12 of 17

multi_galaxy package: new regime package in autolens_workspaceπŸ“– docsautolenslargesupervisedhigh

Profile and speed up JAX likelihood-function compile times (all use cases)✨ featureautolens_profilinglargesupervisedhigh

Give the Profiling Agent a compile-time axis β€” the arc✨ featureprofilinglargesupervisedhigh

+

Test-performance section on the Heart board β€” run times, hangs, NO_RUNβ€¦βœ¨ featurepyautoheartlargesupervisedhigh

Deep research: Can we speed up Delaunay in PyAutoArray?πŸ”¬ researchautoarraytoo-largesupervisedhigh

Census of priors and messages β€” confirmed bugs + redesign ideasπŸ”¬ researchautofittoo-largesupervisedhigh

einstein_radius_jit_from: replace static init_guess with a JAX-native seed finder♻️ refactorautogalaxytoo-largesupervisedhigh

Split lensing regimes: multi_galaxy / group / cluster (epic plan)πŸ“– docsautolenstoo-largesupervisedhigh

-

Re-baseline the MGE imaging JIT profiling regression valueπŸ§ͺ testautolens_workspace_developertoo-largesupervisedhigh

Quick wins (small enough, and safe enough to run unattended)

Defer the eager scipy.sparse import in derivative_util (~0.10 s of import)🧹 maintenancelibrariessmallsafenormal

Bug: fix the tracer.fits existence guard in autolens_workspace imaging modeling.pyπŸ› bugworkspacessmallsafenormal

@@ -163,14 +163,15 @@

Planned

latent-nan-guard-honest-run β€” planned 2026-07-22

Backlog markdown version

-

155 filed prompts, not started β€” sorted most-pickable first (priority, then size). 25 of them belong to an epic and are listed only under Epics below.

+

156 filed prompts, not started β€” sorted most-pickable first (priority, then size). 25 of them belong to an epic and are listed only under Epics below.

-feature β€” 29 +feature β€” 30

Numba CPU likelihood phase 1: batched MGE convolution + operated-matrix caching✨ featureautoarraymediumsupervisedhigh

Rectangular mesh split: Bilinear (fast CPU default) vs RTU (advanced/GPU)✨ featureautoarraymediumsupervisedhigh

Numba CPU likelihood phase 2: kernel-CDF numba fast path (the 49-88% lever)✨ featureautoarraylargesupervisedhigh

Profile and speed up JAX likelihood-function compile times (all use cases)✨ featureautolens_profilinglargesupervisedhigh

Give the Profiling Agent a compile-time axis β€” the arc✨ featureprofilinglargesupervisedhigh

+

Test-performance section on the Heart board β€” run times, hangs, NO_RUNβ€¦βœ¨ featurepyautoheartlargesupervisedhigh

Which other searches need prior-support handling β€” coverage audit after Prodigy✨ featureautofitmediumsupervisedmedium

Give PyAutoFit searches a seed β€” today no search can be madeβ€¦βœ¨ featureautofitmediumsupervisedmedium

Brain board follow-ups: what real mornings surface✨ featurepyautobrainsmallsupervisednormal

@@ -325,6 +326,12 @@

Backlog The 50 newest things to happen to the work in hand, newest first β€” issued, parked, filed. Every other section on this page is laid out by state, which is exactly why none of them can answer β€œwhat has been happening?”. Shipped work is not here: it is read from complete/index.md, and a thousand records deep it would crowd out everything anyone can still act on. Showing the newest 10; … opens the next 10.

+ + + + + + @@ -378,7 +385,7 @@

Backlog Remove pynufft + legacy TransformerNUFFTPyNUFFT

- + @@ -618,12 +625,6 @@

Backlog autolens_workspace_developer: broad stale-API rot (56 symbols, no CI)

- - - - - -
2026-08-24filedTest-performance section on the Heart board β€” run times, hangs…
2026-08-23 filed pynufft removal: unswept downstream residue (1 hard break + stale…

Epics markdown version

diff --git a/dashboard.md b/dashboard.md index 26e04989..33518992 100644 --- a/dashboard.md +++ b/dashboard.md @@ -11,11 +11,11 @@ Every task the Mind is holding, on one page: what is in flight, what is parked, | [In flight](#in-flight) (`active/`) | 1 | | [Parked](#parked) (`parked.md`) | 3 | | [Planned](#planned) (`planned.md`) | 6 | -| [Backlog](#backlog) (`draft/`) | 155 | +| [Backlog](#backlog) (`draft/`) | 156 | ## Start here -**Highest priority** (filed as `high`) β€” showing 12 of 17 +**Highest priority** (filed as `high`) β€” showing 12 of 18
πŸ“‹ TRIAGE: needs manual review before routing β€” medium Β· safe Β· high @@ -73,6 +73,14 @@ Every task the Mind is holding, on one page: what is in flight, what is parked,
+
πŸ“‹ Test-performance section on the Heart board β€” run times, hangs, NO_RUN… β€” pyautoheart Β· large Β· supervised Β· high + +``` +/start_dev draft/feature/pyautoheart/test_performance_board.md +``` + +
+
πŸ“‹ Deep research: Can we speed up Delaunay in PyAutoArray? β€” autoarray Β· too-large Β· supervised Β· high ``` @@ -105,14 +113,6 @@ Every task the Mind is holding, on one page: what is in flight, what is parked,
-
πŸ“‹ Re-baseline the MGE imaging JIT profiling regression value β€” autolens_workspace_developer Β· too-large Β· supervised Β· high - -``` -/start_dev draft/test/autolens_workspace_developer/mge_jit_regression_rebaseline.md -``` - -
- **Quick wins** (small enough, and safe enough to run unattended)
πŸ“‹ Defer the eager scipy.sparse import in derivative_util (~0.10 s of import) β€” libraries Β· small Β· safe Β· normal @@ -235,10 +235,10 @@ Scoped but not started; some are not yet prompt files. Full detail in [`planned. ## Backlog -**155** filed prompts, not started. Each section is sorted most-pickable first (priority, then size). **25** of them belong to an epic and are listed only under [Epics](#epics) below. +**156** filed prompts, not started. Each section is sorted most-pickable first (priority, then size). **25** of them belong to an epic and are listed only under [Epics](#epics) below.
-feature β€” 29 +feature β€” 30
πŸ“‹ Numba CPU likelihood phase 1: batched MGE convolution + operated-matrix caching β€” autoarray Β· medium Β· supervised Β· high @@ -280,6 +280,14 @@ Scoped but not started; some are not yet prompt files. Full detail in [`planned.
+
πŸ“‹ Test-performance section on the Heart board β€” run times, hangs, NO_RUN… β€” pyautoheart Β· large Β· supervised Β· high + +``` +/start_dev draft/feature/pyautoheart/test_performance_board.md +``` + +
+
πŸ“‹ Which other searches need prior-support handling β€” coverage audit after Prodigy β€” autofit Β· medium Β· supervised Β· medium ``` @@ -1328,6 +1336,7 @@ The 50 newest things to happen to the work in hand, newest first β€” issued, par | Date | Event | Task | |------|-------|------| +| 2026-08-24 | filed | Test-performance section on the Heart board β€” run times, hangs… | | 2026-08-23 | filed | pynufft removal: unswept downstream residue (1 hard break + stale… | | 2026-08-23 | filed | Properly time and profile the smoke/release script surface | | 2026-08-23 | filed | Phase 3: stop installing pynufft in Hands/Heart CI and PyAutoCTI… | @@ -1337,12 +1346,12 @@ The 50 newest things to happen to the work in hand, newest first β€” issued, par | 2026-08-22 | filed | Untrack the generated FITS test artifacts in autoarray | | 2026-08-22 | filed | The reconstruction noise map describes a different estimator than the… | | 2026-08-22 | filed | Remove pynufft + legacy TransformerNUFFTPyNUFFT | -| 2026-08-22 | filed | Point-source JSON datasets record no resolution regime |
… 10 more (40 left) | Date | Event | Task | |------|-------|------| +| 2026-08-22 | filed | Point-source JSON datasets record no resolution regime | | 2026-08-22 | filed | Is Intel macOS a supported platform, and what is the numpy-only… | | 2026-08-22 | filed | Defer the eager scipy.sparse import in derivative_util (~0.10 s of… | | 2026-08-22 | filed | Bug: fix the tracer.fits existence guard in autolens_workspace… | @@ -1352,12 +1361,12 @@ The 50 newest things to happen to the work in hand, newest first β€” issued, par | 2026-08-20 | filed | Numba CPU likelihood phase 1: batched MGE convolution +… | | 2026-08-19 | filed | status.sh --repos sources a file that no longer exists | | 2026-08-19 | filed | jax 0.11 breaks beta/gamma message log_partition under jit… | -| 2026-08-19 | filed | autolens_workspace_test jax_likelihood pins: 4 scripts fail smoke on… |
… 10 more (30 left) | Date | Event | Task | |------|-------|------| +| 2026-08-19 | filed | autolens_workspace_test jax_likelihood pins: 4 scripts fail smoke on… | | 2026-08-19 | filed | autofit_profiling: bootstrap the repo + general PyAutoFit profiling… | | 2026-08-19 | filed | autoreduce 0.9 on PyPI never got the Python 3.12 floor | | 2026-08-19 | issued | @PyAutoFit TransformedMessage.factor_gradient crashes on first… | @@ -1367,12 +1376,12 @@ The 50 newest things to happen to the work in hand, newest first β€” issued, par | 2026-08-19 | filed | PyAutoConf rename leftovers in Brain functional surfaces | | 2026-08-19 | filed | Explore: dashboardify the Brain's operational surfaces with pasteable… | | 2026-08-19 | filed | Deduplicate repos_sync.py's check/write pairs | -| 2026-08-19 | filed | Bug in autocti_workspace: the dataset_1d results/database example… |
… 10 more (20 left) | Date | Event | Task | |------|-------|------| +| 2026-08-19 | filed | Bug in autocti_workspace: the dataset_1d results/database example… | | 2026-08-18 | parked | single-source-density-design | | 2026-08-18 | parked | prior-message-collapse-design | | 2026-08-18 | filed | @PyAutoFit TransformedMessage.logpdf/pdf omit the transform… | @@ -1382,12 +1391,12 @@ The 50 newest things to happen to the work in hand, newest first β€” issued, par | 2026-08-14 | filed | Three jax_likelihood pins are stale by ~1.24e-4 and fail the smoke… | | 2026-08-09 | found | isothermal-ell-sph-oversampling-at-the-cusp | | 2026-08-08 | parked | pyautoreduce-slacs1430-acs-comparison | -| 2026-08-08 | filed | Regenerate autolens_workspace markdown/ so the MGE pages show… |
… 10 more (10 left) | Date | Event | Task | |------|-------|------| +| 2026-08-08 | filed | Regenerate autolens_workspace markdown/ so the MGE pages show… | | 2026-08-07 | filed | autofit.plot functions accept **kwargs and silently discard them | | 2026-08-07 | filed | Regenerate setup_notebook-drifted notebooks in… | | 2026-08-06 | filed | Triage: Convolver "No blurring_image provided" warning in canonical… | @@ -1397,7 +1406,6 @@ The 50 newest things to happen to the work in hand, newest first β€” issued, par | 2026-08-04 | filed | dataset/imaging/jwst_lw is untracked because the gitignore was never… | | 2026-08-04 | filed | cosmos_web_ring stores boolean masks as float64, wasting ~3.4 MB of… | | 2026-08-04 | filed | autolens_workspace_developer: broad stale-API rot (56 symbols, no CI) | -| 2026-08-04 | filed | aplt.Output stale-API drift in the remaining workspace repos |
diff --git a/docs/pyautoheart/test_performance_board_assessment.md b/docs/pyautoheart/test_performance_board_assessment.md new file mode 100644 index 00000000..951dba38 --- /dev/null +++ b/docs/pyautoheart/test_performance_board_assessment.md @@ -0,0 +1,281 @@ +# The test-performance board β€” scoping assessment + +Date: 2026-08-24. The deep-research write-up for "the speed of AI development +is heavily dependent on the speed of tests": a dashboard that always shows the +run times of the testing/integration infrastructure (PR smoke gates on the +`*_workspace_test` and normal workspaces, HowTo, unit tests, import time), with +history to flag performance drops, kill-timer/hang events surfaced, NO_RUN +scripts listed with their reasons, and one-tap Claude prompts on every row so +"speed this one up" is a paste, not an archaeology session. Actionable prompt: +`draft/feature/pyautoheart/test_performance_board.md`. + +## The problem, with receipts + +- Manual timing archaeology is the current tool. The jax_grad budget work + records that "the diagnosis had to be rebuilt from CI job logs by hand" + (`complete/2026/08/jax-grad-smoke-timeout-budget.md`), and the 2026-08-23 + slow-vs-stall audit hand-scraped `[PASS] β€” s` lines per run. +- The cost of not watching is measured: four `autogalaxy_workspace_test` runs + burned ~24h of runner time at the 6-hour Actions ceiling before the kill + timer existed (`draft/bug/ci/jax_vmap_jit_compile_stall.md`); the + autolens_workspace_test PR gate is ~11m20s wall-clock of which 553s is + scripts, at ~17 runs/week (`draft/test/workspaces/slowest_smoke_gate_scripts.md`, + `draft/test/pyautoheart/smoke_relevance_gate.md`). +- The existing markers are untrustworthy: "**a SLOW marker is not evidence of + slowness**" β€” the first SLOW-marked entry ever measured was wrong by ~50Γ—, + and every 2026-07-14 marker records no timing at all + (`complete/2026/08/jax-compile-stall-slow-vs-stall-audit.md`). +- The tracked signal is broken: Heart's `script_timing` baselines are orphaned + by path-derived slugs (no history accumulated for the moved jax_grad scripts + since 2026-07-24) and every stored history is one value repeated seven times + (`draft/bug/pyautoheart/script_timing_baselines_orphaned_and_window_filled.md` + β€” filed 2026-08-04, **never issued**). + +## What is recommended + +**A "⏱ Performance" section on the Heart board, published in the Heart's +`board.json` (schema v2, additive) with every row carrying its own one-tap +prompt β€” and a headline row on the Brain board consuming it verbatim, exactly +the way `heart_blockers` already works.** Not a seventh board. + +Why Heart, not Brain: + +- The doctrine is explicit and repeated: "**measurement lives in Heart; + hygiene acts**" and "new standing signals (import cost, CLI noise) become + Heart *legs*, not a new repo" + (`PyAutoBrain/agents/conductors/hygiene/AGENTS.md:143-149`). Heart already + owns every timing signal in the organism: `script_timing`, `test_run`, + `import_time`, `unit_test_timing`, `workspace_testmode_timing`. +- The consumption seam already exists and is tested: the Heart board publishes + structured blockers (`{text, severity, repo, repo_url, run_url, prompt}`) + and the Brain board renders them verbatim, never re-deriving the prompt + (`PyAutoBrain/board/_board.py:228-249`, `tests/test_board.py:43-53`). + A `performance` block rides the same contract; schema v2 is additive by + design (`complete/2026/08/actionable-health-board.md`). +- The cadence already lines up: Heart renders at 05:00 UTC, the Brain board + reads it at 05:30 (`PyAutoBrain/.github/workflows/brain_board.yml:20-25`). +- Timing rows are **advisory, never gating** β€” the precedent is the + `import_time` leg ("advisory dashboard section, NOT in the readiness gating + set", `complete/2026/07/import-time-heart-leg.md`). The four-tier + GREEN/STALE/YELLOW/RED verdict is untouched. + +Why not a Brain-board section (considered): the Brain board holds the exact +code to copy β€” `gh_json` (`_board.py:136-150`), the overnight scrape +(`:167-211`), the self-carrying history (`:456-466`) and `sparkline()` β€” but +putting the *measurement* there splits ownership against the doctrine, and the +Brain "owns no state, no health checks". The Brain board's role is the morning +headline: "N gates slowed / 1 hang flagged", chip β†’ the Heart page. That row +can ride `draft/feature/pyautobrain/brain_board_follow_ups.md`. + +## How the timing information gets calculated and populated + +Three data planes, in cost order. "Just scrape recently completed PRs" β€” the +suggestion that seeded this scoping β€” is plane A, it works today, and it is +already proven in production. + +### Plane A β€” workflow wall-clock, scraped from the Actions API (ships first) + +The default `GITHUB_TOKEN` reads other public PyAutoLabs repos' workflow runs: +`PyAutoMind/.github/workflows/morning_health.yml:68-101` already does exactly +this daily against PyAutoHeart/PyAutoBrain/PyAutoHands/PyAutoFit with +`permissions: contents: read` and no PAT. Per run the API serves everything +the board needs: + +- **duration** = `updated_at βˆ’ run_started_at` (also served directly as + `run_duration_ms` by `GET /actions/runs/{id}/timing`; ignore the `billable` + fields β€” they are 0 on public repos). **Trap, already recorded by the Hands + board: use `run_started_at`, never `created_at`** β€” re-attempted runs + otherwise report multi-day durations (`complete/2026/08/release-board.md`). +- **queue delay** = `run_started_at βˆ’ created_at`. +- **conclusion** (`success | failure | cancelled | …`) β€” see the kill-timer + section for disambiguating `cancelled`. +- **PR association**: `event: pull_request` + `head_branch` β€” so "what a + contributor waits on a PR" is directly measurable, per gate, per week. +- **per-job and per-step timings** via `GET /actions/runs/{id}/jobs` β€” matrix + leg identity is in the job `name` (`"pytest (3.12)"`), and the *Run smoke + tests* step is separable from checkout/install overhead. Fan out to `/jobs` + only for runs worth the detail (the slowest of the day, every `cancelled` + one) β€” that keeps the request budget at ~1 call per tracked workflow per + render, trivially inside the 1000 req/hr limit. + +The tracked-workflow list is declared config, not code (the tenant-firewall +rule): a `performance:` block in `PyAutoHeart/config/repos.yaml` naming each +`repo:workflow` pair β€” every workspace's `Smoke Tests` caller, the HowTo +gates, the organ self-test gates (`tests.yml`), Heart's weekly +`workspace-smoke.yml`, `release-integrate.yml`, and the Brain's +`nightly-release.yml`. One channel note: since #122 the validation channels +are split so runs attribute to the **caller** workflow +(`complete/2026/07/split-validation-channels.md`) β€” query the callers. + +What plane A yields per gate: last-N durations, p50/max, conclusion mix, +queue delay, and the two contributor-facing numbers that matter β€” median PR +gate latency and its trend. + +### Plane B β€” per-script timings, promoted from log lines to a standing dataset + +The runner already prints `[PASS] β€” s` per entry and a +`=== Smoke test summary ===` terminal line; today that data evaporates into +job logs (retained ~90 days) and is only recovered by hand-scraping. The +smoke-runner delegation (`complete/2026/08/smoke-runner-delegation.md`) turned +all ten workspace runners into thin shims over `autohands/run_python.py` β€” +**so per-script timing recording is now one PyAutoHands change, not ten +repo sweeps**: extend the existing report machinery (`--report-dir` is +already load-bearing β€” "without it the gate runs to completion and always +exits 0") to emit a `smoke_timings.json` (entry, status, seconds, cap in +force, exit code), uploaded as a run artifact and/or written to +`$GITHUB_STEP_SUMMARY`. The Heart render fetches the latest artifact per gate. + +This is precisely the open question item 4 of +`draft/research/ci/smoke_timing_and_profiling.md` already poses ("should the +runner record per-script timings routinely, so this is a standing dataset +rather than a periodic archaeology exercise?") β€” this assessment's answer is +**yes, via the delegated runner**. The `retime.yml` classifier and its +verdict vocabulary (STALL/SLOW/NEITHER/AMBIGUOUS/ERROR, `retime_results.json`) +stay as the on-demand deep probe, reached through `smoke-tests.yml`'s +`runner`/`runner-args`/`script-timeout` inputs. + +**STALL β‰  SLOW is a first-class dimension, not a footnote.** "A slow script +has a tight timing distribution. A stalling one is bimodal" β€” a healthy +compile of `rectangular_mge.py` is 3.1s; a stalled one exceeds 300s, same +commit, same runner image (`jax-compile-stall-slow-vs-stall-audit.md`). A +single wall-clock number per entry cannot express this; the board should carry +the retime verdict where one exists, and flag bimodality from the standing +dataset where it doesn't. + +### Plane C β€” baselines and regression flagging (the "history" that must be durable) + +Two history mechanisms exist in the board family, with different guarantees: + +1. **Self-carrying published artifact** β€” the Brain board fetches its own + previous `board.json` at render, appends today, caps at 30 entries + (`PyAutoBrain/board/_board.py:456-466`). Free, no commits, idempotent per + date β€” but lossy (a publish gap loses everything) and 30 days max. Right + for the plane-A per-gate daily aggregates and their sparklines. +2. **Heart's tracked rolling baselines** (`~/.pyauto-heart/`, the + `script_timing` mechanism) β€” durable, but currently broken as filed. Right + for per-script baselines once fixed. + +The fix is a prerequisite, and its prompt is already written: +`draft/bug/pyautoheart/script_timing_baselines_orphaned_and_window_filled.md` +(rename-aware slugs or a loud no-baseline signal; real 7-run accumulation; +record the source run id per duration). **Issue it first.** + +For flagging drops, do not invent thresholds β€” reuse the profiling +conductor's considered doctrine: a regression counts only if it is newer than +its pin, **at least 2.0Γ— the pinned value AND at least 1.0s above it in +absolute terms**, with sticky pins that never move without an explicit +`--repin` ("host load alone has produced 7Γ— errors in this corpus and an +alarm that cries wolf gets ignored" β€” +`PyAutoBrain/agents/conductors/profiling/AGENTS.md`). The comparability-key +lesson from the compile-warm dashboard transfers too: never pool different +hosts under one label β€” for CI the key is runner image Γ— Python leg Γ— event +type (`complete/2026/08/compile-warm-baseline-dashboard.md`). + +Unit tests and import time ride the same planes: the organ/library `tests.yml` +gates are plane-A rows (they are seconds-to-a-minute today β€” the board's job +is to notice when that stops being true), and the existing `import_time` leg +(advisory, off-tick, subprocess-measured) surfaces its red/yellow counts as a +row with a `/hygiene` chip β€” the standing leg promotion that +`hygiene/AGENTS.md:117-119` already names as the deferred follow-up. + +## Kill-timer and hang events on the board + +The kill timer (PyAutoHands `build_util.timeout_for` + `kill_group`; 300s +smoke default, 900s `jax_grad/`, 1800s release; exit 124; `TIMEOUT` status +with the truncated output tail and the cap in force attached) and the +in-process watchdog (faulthandler dump at 80% of `BUILD_SCRIPT_TIMEOUT`, +heartbeat lines) already leave ingestible traces. The board renders, per gate: + +- **Per-script TIMEOUT rows** (plane B): entry, cap, count over the window, + and the retime verdict if one exists. Chip: + `/bug kill timer: