record: searches-readme-dashboard shipped (autolens_profiling#139)

active.md

@@ -1,17 +1,5 @@
 # Active Tasks
 
-## searches-readme-dashboard
-- issue: https://github.com/PyAutoLabs/autolens_profiling/issues/138
-- session: Claude Code cloud/mobile (no local worktree)
-- status: library-dev
-- worktree: none — cloud session works on the GitHub clone directly
-- repos:
-  - autolens_profiling: feature/searches-readme-dashboard
-- prompt: active/searches_readme_dashboard_no_data.md
-- note: Phase 0(e) of the inference programme (PROGRAMME.md). Root cause verified:
-  build_readme.py scans a flat filename layout; searches artifacts live in nested
-  cells with bare-tier filenames, and the renderer reads a stale payload schema.
-
 ## stored-sample-reconstruction-guard
 - issue: https://github.com/PyAutoLabs/PyAutoFit/issues/1486
 - status: library-dev — WORKSPACE HALF SHIPPED; the PyAutoFit hardening (#1486) is what remains

active/searches_readme_dashboard_no_data.md → complete/2026/08/searches-readme-dashboard.md

@@ -1,3 +1,38 @@
+Fixed the permanently-empty searches README dashboard in autolens_profiling —
+Phase 0(e) of the inference programme
+(`results/notes/inference/PROGRAMME.md`). Documentation-tooling only; no
+runs, no source-library changes. Done entirely from a mobile/cloud session
+(no local worktree).
+
+- **PR:** autolens_profiling#139 (merged `e43e1f1`, closes #138).
+- **Root cause:** two decoupled stalenesses in
+  `scripts/misc/tooling/build_readme.py`. The scanner only matched flat
+  versioned filenames (`<script>_summary_v<version>.json`) or
+  script-prefixed config tags, while the searches framework writes
+  `results/searches/<sampler>/<class>/<model>/<instrument>/<config>.json`
+  with bare-tier names — nothing matched, so the `searches-nautilus` region
+  always rendered the no-data block. The renderer additionally read a payload
+  schema that never shipped (`performance.wall_time_s`,
+  `convergence.evals_to_ml`, top-level `backend`) instead of the real one
+  (`performance.total_wall_s`/`likelihood_evals`/`time_per_eval_ms`,
+  `results.max_log_likelihood`/`log_evidence`).
+- **Fix:** payload-driven `SearchArtifact` scanner (identity from the
+  self-describing JSON, no path parsing; skips the list-shaped
+  `multi_start_nan_accounting` overhead study), one all-samplers table
+  (latest version per sampler × cell × config), region renamed
+  `searches-nautilus` → `searches`. 34 rows render.
+- **Verification:** nautilus `imaging/mge/hst` row matches the
+  inference-programme truth bar exactly (max logL 31,786.8, logZ 31,690.5,
+  831 s, 63,800 evals); NSS row matches (679 s, 394,321 evals, logZ
+  31,697.7). `build_readme.py --check` (CI gate) and ruff both green.
+- **Trap for the record:** the `--check` CI gate means any renderer change
+  must regenerate all target READMEs in the same PR, or main goes red.
+- **Follow-up:** PROGRAMME.md's phase-state table still lists 0(e) as
+  outstanding — updated in the next inference-notes PR rather than a
+  one-line PR of its own.
+
+## Original prompt
+
 # searches README dashboard renders "No data yet" despite existing artifacts
 
 Type: bug

complete/index.md

@@ -6,7 +6,7 @@ Token-light navigation over the finished-work records (schema:
 only then grep a dated bucket. Curators: edit the band between the CURATED
 markers; everything below GENERATED is rebuilt.
 
-1010 records across 7 buckets.
+1011 records across 7 buckets.
 
 <!-- CURATED:START -->
 ## Highlights
@@ -121,6 +121,7 @@ _(curate hard-won records here — survives regeneration.)_
 - [samplers-surface-autolens-tiers](2026/08/samplers-surface-autolens-tiers.md) — the samplers faculty's SamplerSurface now scans the findings
 - [save-json-numpy-scalar-typeerror](2026/08/save-json-numpy-scalar-typeerror.md) — Adds `NumpyEncoder` in `autofit/tools/util.py` (`np.ndarray` ->
 - [script-title-underline-off-by-one](2026/08/script-title-underline-off-by-one.md)
+- [searches-readme-dashboard](2026/08/searches-readme-dashboard.md)
 - [simulator-util-to-af-ex](2026/08/simulator-util-to-af-ex.md) — moved the four 1D-Gaussian simulator helpers out of the duplicated
 - [small-datasets-loader-pixel-scales](2026/08/small-datasets-loader-pixel-scales.md)
 - [smoke-runner-jupyter-guard](2026/08/smoke-runner-jupyter-guard.md)