diff --git a/.claude/hooks/session-start.sh b/.claude/hooks/session-start.sh
index 54a10fe6e..591c2d7c6 100755
--- a/.claude/hooks/session-start.sh
+++ b/.claude/hooks/session-start.sh
@@ -132,12 +132,12 @@ marked_root() {
 holds_an_organ() {
     [ -d "$1/PyAutoBrain" ] || [ -d "$1/PyAutoMind" ] || [ -d "$1/PyAutoCortex" ] \
         || [ -d "$1/PyAutoMemory" ] || [ -d "$1/PyAutoEyes" ] || [ -d "$1/PyAutoHeart" ] \
-        || [ -d "$1/PyAutoHands" ] || [ -d "$1/PyAutoPulse" ] || [ -d "$1/PyAutoNerves" ] \
+        || [ -d "$1/PyAutoHands" ] || [ -d "$1/PyAutoPulse" ] || [ -d "$1/PyAutoInsight" ] || [ -d "$1/PyAutoNerves" ] \
         || [ -d "$1/PyAutoGut" ] \
         || [ -d "$1/organs/PyAutoBrain" ] || [ -d "$1/organs/PyAutoMind" ] \
         || [ -d "$1/organs/PyAutoCortex" ] || [ -d "$1/organs/PyAutoMemory" ] \
         || [ -d "$1/organs/PyAutoEyes" ] || [ -d "$1/organs/PyAutoHeart" ] \
-        || [ -d "$1/organs/PyAutoHands" ] || [ -d "$1/organs/PyAutoPulse" ] \
+        || [ -d "$1/organs/PyAutoHands" ] || [ -d "$1/organs/PyAutoPulse" ] || [ -d "$1/organs/PyAutoInsight" ] \
         || [ -d "$1/organs/PyAutoNerves" ] || [ -d "$1/organs/PyAutoGut" ]
 }
 
diff --git a/AGENTS.md b/AGENTS.md
index b318073a5..641f9d738 100644
--- a/AGENTS.md
+++ b/AGENTS.md
@@ -6,6 +6,11 @@ when the task needs them. Read [README.md](README.md) for the public overview.
 
 ## Where work lives
 
+- Inference campaign intent and pending domain tasks live in **PyAutoInsight**
+  (`campaigns.yaml`, `tasks/`, `CHECKIN.md`). Mind retains bounded implementation
+  issue/PR lifecycle and claims. See `docs/insight-task-migration.md`; source
+  removal follows the destination merge, never precedes it.
+
 - Profiling campaign intent and pending tasks live in **PyAutoPulse**
   (`campaigns.yaml`, `tasks/`, `CHECKIN.md`). Use its single-chat check-in first.
   Mind holds bounded implementation phases and repository claims, linking the
diff --git a/active.md b/active.md
index 16d9d891a..975c80d40 100644
--- a/active.md
+++ b/active.md
@@ -37,3 +37,29 @@
 - workspace-pr: https://github.com/PyAutoLabs/autolens_workspace_test/pull/343
 - workspace-pr: https://github.com/PyAutoLabs/autolens_profiling/pull/365
 - final-validation: Galaxy 1315 passed; focused 49 passed; full companion smoke 33/33 in 501.91s (changed example 2.2s). Galaxy exact-head CI all green; new companion CI pending. Wiki/results-layout and whitespace checks pass. No raw phase-3a evidence changed.
+
+## insight-organ-birth
+- issue: https://github.com/PyAutoLabs/PyAutoInsight/issues/1
+- issued: 2026-10-04
+- prompt: active/insight_organ_birth.md
+- session: Codex Work
+- status: workspace-dev
+- repos:
+  - PyAutoInsight: feature/insight-control-room
+  - autolens_inference: feature/insight-producer
+  - PyAutoMind: feature/insight-integration
+  - PyAutoBrain: feature/insight-integration
+  - PyAutoHeart: feature/insight-integration
+  - PyAutoHands: feature/insight-integration
+  - PyAutoCortex: feature/insight-integration
+  - PyAutoPulse: feature/insight-integration
+  - PyAutoMemory: feature/insight-integration
+  - PyAutoEyes: feature/insight-integration
+  - PyAutoEars: feature/insight-integration
+  - PyAutoNerves: feature/insight-integration
+  - PyAutoGut: feature/insight-integration
+  - PyAutoScientist: feature/insight-integration
+  - pyautolabs.github.io: feature/insight-integration
+  - .github: feature/insight-integration
+- resume: Producer PR18; Brain460; hub28; .github31; Scientist41; Cortex56; Pulse8; Memory114; Eyes15; Ears8; Heart273; Hands296; Nerves183; Gut20. Insight control-room PR being published. Wait human /prm on green; source cleanup only after Insight destination merges. No compute launched.
+- heart-red-override: Human 2026-10-04 "I authorize on red and repo made, continue"; applies to Insight#1 development; full test/CI and merge gates retained. Exact published reason set in active/insight_organ_birth.md.
diff --git a/active/insight_organ_birth.md b/active/insight_organ_birth.md
new file mode 100644
index 000000000..ad5451c33
--- /dev/null
+++ b/active/insight_organ_birth.md
@@ -0,0 +1,103 @@
+# PyAutoInsight: inference campaigns and evidence organ
+
+Type: feature
+Target: @PyAutoInsight
+Repos:
+- PyAutoInsight
+- autolens_inference
+- PyAutoMind
+- PyAutoBrain
+Consequence: judge
+Autonomy: supervised
+Priority: high
+Status: active
+Filed: 2026-10-04
+Issued: 2026-10-04
+Issue: https://github.com/PyAutoLabs/PyAutoInsight/issues/1
+
+## Human request
+
+"Build the inference organ end to end, following everything we implemented for PyAutoPulse on 2–3 October 2026, including its latest campaign control room and task migration. This is an implementation request, not just a scoping exercise."
+
+"The outcome should be one dashboard and one ongoing chat through which I can manage all inference campaigns, see updates, direct individual campaigns and suggest ideas."
+
+Name chosen: PyAutoInsight, organ key insight. User created public repository.
+Current Heart override: "I authorize on red and repo made, continue" (2026-10-04).
+
+## Delivery contract
+
+Follow Insight#1. Preserve Pulse's current control-room design/order: editable
+copyable all-campaign prompt; active campaign table; open tasks; evidence below.
+Project-owned inference-summary@1 exporter reads actual results including failed,
+stopped, incomplete, archived and diagnostic-missing runs, with exact provenance,
+parent/stage identity and separately defined timing clocks. Never sum overlapping
+stages or infer scientific success from process completion. Raw samples remain
+project-owned and availability unknown unless verified.
+
+Insight owns registry resolved through Mind identities, schema reader, one-commit
+ingest receipts, last-good snapshots, campaign/task intent and durable check-ins.
+Cortex owns science runs/observations/human conclusions. Mind retains bounded PR
+lifecycle and claims. No new inference conductor or retired programme evidence.
+
+Audit all pending Mind sources; migration preserves bytes/hashes/source commit,
+priorities/decisions/blockers/issues and duplicate scheduling removal. Destination
+must merge before Mind source removal. Regenerate canonical body maps/adapters,
+Brain cockpit, public hub/org lists. Include producer dispatch sender and verify
+published receipt against successful publication revision. Full CI, tests, review
+and required authorisations before merge. No compute or release authorised.
+
+## Validation / resumability
+
+Implementation in feature/insight-producer, feature/insight-control-room and
+feature/insight-integration across affected repos. Missing scientific local deps
+must be covered by exact-head full CI, never a fabricated local pass.
+
+## Exact published Heart RED reason set acknowledged at entry
+
+```text
+release validation FAILED (stage integrate)
+PyAutoHeart/workflows/Release Integrate: failure
+PyAutoArray/Tests: red
+PyAutoArray/Tests: red
+PyAutoFit/Tests: red
+autofit_workspace_test/Smoke Tests: red
+autogalaxy_workspace_test/Smoke Tests: red
+autolens_workspace_test/Smoke Tests: red
+CI wall-clock: 26 gates · slowest PyAutoBrain Nightly Release 83m · 1 slowed · 6 hang events
+Release readiness: release validation FAILED (stage integrate)
+PyAutoCTI/test_autocti/extract/two_d/parallel/test_parallel_fpr.py::test__estimate_capture: red
+PyAutoNerves/test_autonerves/test_fitsable.py::test__output_to_fits: red
+Unit-test timing: 3 test regressions (>3× baseline), 1 slow (>1.5×)
+autolens_test: red
+Release validation: NOT release_ready — v2026.10.3.1.dev79901 profile=release (2026-10-03T07:43:31+00:00)
+validation_report/stages/integrate: fail
+autolens_assistant: red
+autolens_profiling: red
+autolens_workspace_test: red
+Worktree drift: 0 orphan / 0 missing / 4 dirty
+euclid_strong_lens_modeling_pipeline: red
+```
+
+## Shipping checkpoint — 2026-10-04
+
+- Organ: https://github.com/PyAutoLabs/PyAutoInsight/pull/2; 77 tests,
+  lint and hosted Dashboard Refresh pass. No check-in stamped by refresh.
+- Producer: https://github.com/PyAutoLabs/autolens_inference/pull/18;
+  full hosted lint workflow passes, including scientific simulator/SLaM smoke.
+- Registration: https://github.com/PyAutoLabs/PyAutoMind/pull/471.
+- Integration: Brain #460, hub #28, .github #31, Scientist #41,
+  Cortex #56, Pulse #8, Memory #114, Eyes #15, Ears #8,
+  Heart #273, Hands #296, Nerves #183, Gut #20.
+  All triggered integration PR workflows passed at their published heads.
+- Destination contains eight campaigns and six frozen source task records;
+  Mind source originals remain until the destination is merged.
+- Captured producer evidence: feature/insight-producer commit
+  9c362a7a298238d766f61e812c694813e94830d1 (56 records, 15 parent runs).
+  This is explicitly branch evidence, not a verified main publication.
+- Remaining gated sequence: explicit human merge command under AUTONOMY.md
+  RED development override; merge destination/registration/integrations and
+  producer; hash-checked Mind source cleanup and generated ledger round-trip;
+  verify actual producer publication dispatch, expected receipt revision and
+  published GitHub Pages; then lifecycle close-out. No merge or deployment
+  verification has been claimed. Browser visual check remains outstanding;
+  editable-copy native/fallback behavior is covered by executable Node tests.
diff --git a/autonomy_log.md b/autonomy_log.md
index 27e12e69d..1214d6c66 100644
--- a/autonomy_log.md
+++ b/autonomy_log.md
@@ -9,6 +9,7 @@ Outcome ∈ `merged-unchanged` / `amended` / `rejected` / `parked` /
 
 | date | task | effective level | gates (tests/smoke/review/heart) | outcome |
 |------|------|-----------------|----------------------------------|---------|
+| 2026-10-04 | insight-organ-birth (Insight#1) | supervised | Validation in progress; current task-specific human "I authorize on red and repo made, continue"; exact reasons in task; no compute/release/CI bypass | red-override |
 | 2026-10-03 | ears-integration (Brain#455) | supervised | Brain 1150 / Mind 619 pass; review CLEAN; live feed + copy verified; Heart RED reasons in active.md and task issue; accepted whole-task development authorization resumed by user; no merge/release | red-override |
 | 2026-10-01 | ecosystem-routing-trial organ specification (Brain#444 / PR#445) | supervised | Sphinx zero warnings; seven pinned citations and whitespace PASS; frozen trial unchanged; author contract review (no new Fable review); scientific smoke N/A; Heart RED `autofit_workspace: Smoke Tests failure on main`; live user "I authorize, continue" for task-specific documentation shipping; commit/push/PR update only, no merge/release/CI bypass | red-override |
 | 2026-10-01 | ecosystem-layers (PyAutoBrain#440 / PR #441, 9ae0db9) | supervised | Sphinx zero warnings; 32 citations/17 pinned sources; whitespace PASS; independent Claude Fable 5.1 draft review CLEAN; scientific smoke N/A (research prose only); Heart RED `release validation FAILED (stage integrate)`; live user "yes do it", then "I authorize", to task-specific development override; commit/push/PR only, no merge/release/rehearsal/CI bypass | red-override |
diff --git a/dashboard.html b/dashboard.html
index e1fda0036..c569ea7cb 100644
--- a/dashboard.html
+++ b/dashboard.html
@@ -4,7 +4,7 @@
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
 <title>PyAutoMind Dashboard</title>
-<!-- generated by `pyauto-brain intake dashboard --apply` on 2026-10-03 — regenerate, do not hand-edit -->
+<!-- generated by `pyauto-brain intake dashboard --apply` on 2026-10-04 — regenerate, do not hand-edit -->
 <style>:root{color-scheme:light dark;--bg:#fff;--fg:#1f2328;--muted:#59636e;
  --line:#d8dee4;--btn:#f6f8fa;--ok:#1a7f37;--warn:#9a6700;--bad:#d1242f;
  --accent:#0a7d72;--tint:#0a7d7214;--edge:#0a7d723d;
@@ -231,8 +231,8 @@
 </head>
 <body>
 <header class="hero"><span class="orb"><svg class="mark" viewBox="0 0 48 48" fill="none" stroke="currentColor" stroke-width="1.3" stroke-linecap="round" stroke-linejoin="round" aria-hidden="true"><circle cx="24" cy="24" r="20.4"/><path d="M20.2,37.6 L20.2,33.6 C20.2,32.4 19.4,31.6 18.4,30.8 C16.8,29.6 15.8,28.2 15.6,26.6 C15.5,25.6 14.9,25.2 13.9,24.9 C12.6,24.5 12.3,23.7 13.0,22.7 C13.8,21.6 14.5,20.6 14.6,19.4 C15.0,15.4 18.2,12.2 22.4,11.8 C27.2,11.3 31.6,14.6 32.3,19.2 C32.8,22.4 31.6,25.0 30.6,27.0 C29.9,28.4 29.6,29.6 29.6,31.2 L29.6,37.6"/><circle cx="24.6" cy="19.6" r="4.4"/><path d="M22.5,19.7 L24.0,21.3 L26.8,18.1"/><path d="M34.2,16.5 L38.8,16.5 M35.2,21.0 L39.4,21.0 M33.6,25.5 L37.2,25.5"/><g fill="currentColor" stroke="none"><circle cx="40.1" cy="16.5" r="1.4"/><circle cx="40.7" cy="21.0" r="1.4"/><circle cx="38.5" cy="25.5" r="1.4"/></g></svg></span><h1>PyAuto<b>Mind</b><span class="kind">Dashboard</span></h1><div class="rule"></div><p class="tag">Intent. Priority. Flow.</p></header><p class="lede">Every task the Mind is holding. Tap a task's 📋 and its start-dev skill prompt is on your clipboard — paste it into an AI assistant chat to route the assistant straight to that task. <a href="#recent">Recent</a> is the same work by date — what has been happening rather than what to do next.</p>
-<ul class="stats"><li><b>2</b><span>In flight</span></li><li><b>2</b><span>Human review</span></li><li><b>6</b><span>Parked</span></li><li><b>7</b><span>Planned</span></li><li><b>232</b><span>Backlog</span></li></ul>
-<div class="fresh"><p><b>Last updated 2026-10-03.</b> This page is generated from <code>active/</code>, <code>draft/</code> and the registry files, so it is only as current as they are. <code>dashboard_refresh.yml</code> re-renders it on every push to <code>main</code> — that heals a stale page, but not a stale prompt: a task that shipped without its prompt advancing to <code>complete/</code> keeps rendering here as pickable backlog. Reconciling those is the refresh below.</p>
+<ul class="stats"><li><b>3</b><span>In flight</span></li><li><b>2</b><span>Human review</span></li><li><b>6</b><span>Parked</span></li><li><b>7</b><span>Planned</span></li><li><b>232</b><span>Backlog</span></li></ul>
+<div class="fresh"><p><b>Last updated 2026-10-04.</b> This page is generated from <code>active/</code>, <code>draft/</code> and the registry files, so it is only as current as they are. <code>dashboard_refresh.yml</code> re-renders it on every push to <code>main</code> — that heals a stale page, but not a stale prompt: a task that shipped without its prompt advancing to <code>complete/</code> keeps rendering here as pickable backlog. Reconciling those is the refresh below.</p>
 <div class="task"><button class="copy" data-cmd="Bring the PyAutoMind dashboard up to date. Work in the PyAutoMind checkout:
 
 1. `git fetch origin &amp;&amp; git status`. If behind `origin/main`, `git pull --ff-only`
@@ -291,6 +291,7 @@ <h3>Fits a slot <span class="facets">(ready to run unattended, cheapest to revie
 <h2>In flight <a class="mdsrc" href="https://github.com/PyAutoLabs/PyAutoMind/blob/main/active.md">markdown version</a></h2>
 <p class="muted">Issued — each has an open GitHub issue and usually a branch.</p>
 <div class="task"><button class="copy" data-cmd="Use the start-dev skill. active/evaluation_grid_cap_preserves_field.md" aria-label="Copy the AI command">📋</button><p><a href="https://github.com/PyAutoLabs/PyAutoMind/blob/main/active/evaluation_grid_cap_preserves_field.md">Preserve the LensCalc evaluation field when the grid cap activates</a> — <a href="https://github.com/PyAutoLabs/PyAutoGalaxy/issues/645">issue #645</a><span class="facets"> — issued 2026-10-02</span> — <span class="facets">library-merged, workspace-release-gated</span> — PRs: <a href="https://github.com/PyAutoLabs/PyAutoGalaxy/pull/646">PyAutoGalaxy#646</a>, <a href="https://github.com/PyAutoLabs/autolens_workspace_test/pull/343">autolens_workspace_test#343</a>, <a href="https://github.com/PyAutoLabs/autolens_profiling/pull/365">autolens_profiling#365</a> — ⏳ pending release: PyAutoGalaxy</p></div>
+<div class="task"><button class="copy" data-cmd="Use the start-dev skill. active/insight_organ_birth.md" aria-label="Copy the AI command">📋</button><p><a href="https://github.com/PyAutoLabs/PyAutoMind/blob/main/active/insight_organ_birth.md">PyAutoInsight: inference campaigns and evidence organ</a> — <a href="https://github.com/PyAutoLabs/PyAutoInsight/issues/1">issue #1</a><span class="facets"> — issued 2026-10-04</span> — <span class="facets">workspace-dev</span></p></div>
 <div class="task"><button class="copy" data-cmd="Use the start-dev skill. active/vis_lp_inspection_bundle.md" aria-label="Copy the AI command">📋</button><p><a href="https://github.com/PyAutoLabs/PyAutoMind/blob/main/active/vis_lp_inspection_bundle.md">Support vis_lp-only products in inspection bundles</a><span class="facets"> — issued 2026-09-22</span></p></div>
 <a id="pending-release"></a><h2>Pending release <a class="mdsrc" href="https://github.com/PyAutoLabs/PyAutoMind/blob/main/active.md">markdown version</a></h2>
 <p class="muted">Library PRs the ledger records as merged but not yet released, and the in-flight tasks waiting on each. Rendered from the ledger — <code>active.md</code> and the <code>complete/</code> records — never a live GitHub query; the Brain board's <code>pending-release</code> search is the fresh view, this is what the Mind believes.</p>
@@ -967,6 +968,12 @@ <h2>Bundles <a class="mdsrc" href="https://github.com/PyAutoLabs/PyAutoMind/blob
 <p class="muted">The 50 newest things to happen to the work in hand, newest first — issued, parked, filed. Every other section on this page is laid out by state, which is exactly why none of them can answer “what has been happening?”. Shipped work is not here: it is read from <code>complete/index.md</code>, and a thousand records deep it would crowd out everything anyone can still act on. Showing the newest 10; … opens the next 10.</p>
 <table class="recent">
 <tr>
+<td class="when">2026-10-04</td>
+<td class="what">issued</td>
+<td><a href="https://github.com/PyAutoLabs/PyAutoMind/blob/main/active/insight_organ_birth.md">PyAutoInsight: inference campaigns and evidence organ</a></td>
+<td class="pick"><button class="copy" data-cmd="Use the start-dev skill. active/insight_organ_birth.md" aria-label="Copy the AI prompt">📋</button></td>
+</tr>
+<tr>
 <td class="when">2026-10-02</td>
 <td class="what">filed</td>
 <td><a href="https://github.com/PyAutoLabs/PyAutoMind/blob/main/planned.md#sibson-internal-edge-continuity">sibson-internal-edge-continuity</a></td>
@@ -1020,7 +1027,7 @@ <h2>Bundles <a class="mdsrc" href="https://github.com/PyAutoLabs/PyAutoMind/blob
 <td><a href="https://github.com/PyAutoLabs/PyAutoMind/blob/main/draft/docs/autolens_inference/reconcile_cortex_ledger_references.md">Reconcile inference documentation with the current Cortex ledger…</a></td>
 <td class="pick"><button class="copy" data-cmd="Use the start-dev skill. draft/docs/autolens_inference/reconcile_cortex_ledger_references.md" aria-label="Copy the AI prompt">📋</button></td>
 </tr>
-<tr>
+<tr hidden>
 <td class="when">2026-10-01</td>
 <td class="what">filed</td>
 <td><a href="https://github.com/PyAutoLabs/PyAutoMind/blob/main/draft/bug/autoarray/interferometer_log_evidence_large_regularization_coefficient_cholesky.md">Interferometer log_evidence fails (LinAlgError / NaN) at large…</a></td>
@@ -1260,12 +1267,6 @@ <h2>Bundles <a class="mdsrc" href="https://github.com/PyAutoLabs/PyAutoMind/blob
 <td><a href="https://github.com/PyAutoLabs/PyAutoMind/blob/main/draft/bug/autoarray/adaptive_mesh_degenerate_pixel_pair.md">Adaptive Delaunay mesh places two mesh pixels 0.001" apart — a…</a></td>
 <td class="pick"><button class="copy" data-cmd="Use the start-dev skill. draft/bug/autoarray/adaptive_mesh_degenerate_pixel_pair.md" aria-label="Copy the AI prompt">📋</button></td>
 </tr>
-<tr hidden>
-<td class="when">2026-09-15</td>
-<td class="what">filed</td>
-<td><a href="https://github.com/PyAutoLabs/PyAutoMind/blob/main/draft/maintenance/howtofit/minimum_library_version_stale.md">The workspace staleness warning keys off a value documented never to…</a></td>
-<td class="pick"><button class="copy" data-cmd="Use the start-dev skill. draft/maintenance/howtofit/minimum_library_version_stale.md" aria-label="Copy the AI prompt">📋</button></td>
-</tr>
 </table>
 <button class="more" data-page="10">… 10 more (40 left)</button>
 <h2>Epics <a class="mdsrc" href="https://github.com/PyAutoLabs/PyAutoMind/blob/main/epics.md">markdown version</a></h2>
diff --git a/dashboard.md b/dashboard.md
index d0a004c2f..7e6278a23 100644
--- a/dashboard.md
+++ b/dashboard.md
@@ -1,12 +1,12 @@
 # PyAutoMind Dashboard
 
-<!-- generated by `pyauto-brain intake dashboard --apply` on 2026-10-03 — regenerate, do not hand-edit -->
+<!-- generated by `pyauto-brain intake dashboard --apply` on 2026-10-04 — regenerate, do not hand-edit -->
 
 This is the markdown version of the [PyAutoMind Dashboard](https://pyautolabs.github.io/PyAutoMind/), which puts a task's command on your clipboard with a single tap of 📋.
 
 Every task the Mind is holding, on one page: what is in flight, what is parked, and the whole backlog to pick from. Pick a task and run its start-dev skill prompt in an AI assistant chat to start it. [Recent](#recent) is the same work by date — what has been happening rather than what to do next.
 
-> **Last updated 2026-10-03.** This page is generated from `active/`, `draft/` and the registry files, so it is only as current as they are. `dashboard_refresh.yml` re-renders it on every push to `main` — that heals a stale page, but not a stale prompt: a task that shipped without its prompt advancing to `complete/` keeps rendering here as pickable backlog. Reconciling those is the refresh below.
+> **Last updated 2026-10-04.** This page is generated from `active/`, `draft/` and the registry files, so it is only as current as they are. `dashboard_refresh.yml` re-renders it on every push to `main` — that heals a stale page, but not a stale prompt: a task that shipped without its prompt advancing to `complete/` keeps rendering here as pickable backlog. Reconciling those is the refresh below.
 
 <details><summary>📋 <b>Refresh this page</b> — reconcile finished prompts, then regenerate</summary>
 
@@ -42,7 +42,7 @@ anything you could not verify.
 
 | Where | Count |
 |-------|------:|
-| [In flight](#in-flight) (`active/`) | 2 |
+| [In flight](#in-flight) (`active/`) | 3 |
 | [Human review](#human-review) (`draft/human_review/`) | 2 |
 | [Parked](#parked) (`parked.md`) | 6 |
 | [Planned](#planned) (`planned.md`) | 7 |
@@ -260,6 +260,14 @@ Use the start-dev skill. active/evaluation_grid_cap_preserves_field.md
 
 </details>
 
+<details><summary>📋 <a href="active/insight_organ_birth.md">PyAutoInsight: inference campaigns and evidence organ</a> — <a href="https://github.com/PyAutoLabs/PyAutoInsight/issues/1">issue #1</a> — issued 2026-10-04 — workspace-dev</summary>
+
+```
+Use the start-dev skill. active/insight_organ_birth.md
+```
+
+</details>
+
 <details><summary>📋 <a href="active/vis_lp_inspection_bundle.md">Support vis_lp-only products in inspection bundles</a> — issued 2026-09-22</summary>
 
 ```
@@ -2456,6 +2464,7 @@ The 50 newest things to happen to the work in hand, newest first — issued, par
 
 | Date | Event | Task |
 |------|-------|------|
+| 2026-10-04 | issued | <a href="active/insight_organ_birth.md">PyAutoInsight: inference campaigns and evidence organ</a> |
 | 2026-10-02 | filed | <a href="planned.md#sibson-internal-edge-continuity">sibson-internal-edge-continuity</a> |
 | 2026-10-02 | filed | <a href="planned.md#knn-partial-point-block">knn-partial-point-block</a> |
 | 2026-10-02 | parked | <a href="parked.md#colab-refinement-throughout">colab-refinement-throughout</a> |
@@ -2465,12 +2474,12 @@ The 50 newest things to happen to the work in hand, newest first — issued, par
 | 2026-10-02 | filed | <a href="draft/bug/autoarray/fix_knn_neighbor_search_for_a_partial.md">Fix KNN neighbor search for a partial final point block</a> |
 | 2026-10-01 | filed | <a href="draft/bug/autoarray/sparse_terms_nufft_origin_and_mask_compatibility.md">Sparse interferometer terms: NUFFT ignores the mask origin and…</a> |
 | 2026-10-01 | filed | <a href="draft/docs/autolens_inference/reconcile_cortex_ledger_references.md">Reconcile inference documentation with the current Cortex ledger…</a> |
-| 2026-10-01 | filed | <a href="draft/bug/autoarray/interferometer_log_evidence_large_regularization_coefficient_cholesky.md">Interferometer log_evidence fails (LinAlgError / NaN) at large…</a> |
 
 <details><summary>… 10 more (40 left)</summary>
 
 | Date | Event | Task |
 |------|-------|------|
+| 2026-10-01 | filed | <a href="draft/bug/autoarray/interferometer_log_evidence_large_regularization_coefficient_cholesky.md">Interferometer log_evidence fails (LinAlgError / NaN) at large…</a> |
 | 2026-10-01 | filed | <a href="draft/bug/autogalaxy/galaxy_image_dict_mixed_galaxy_overwrites_ordinary_light.md">Interferometer <code>galaxy_image_dict</code> drops a mixed galaxy's ordinary…</a> |
 | 2026-09-29 | filed | <a href="draft/bug/autogalaxy/ellipse_plotter_fit_ellipse_overwrites.md"><code>PlotterEllipse.fit_ellipse</code> writes every variant to…</a> |
 | 2026-09-29 | filed | <a href="draft/maintenance/pyautomind/session_start_hook_copies_regen.md">Regenerate the 28 drifted <code>.claude/hooks/session-start.sh</code> copies…</a> |
@@ -2480,12 +2489,12 @@ The 50 newest things to happen to the work in hand, newest first — issued, par
 | 2026-09-27 | filed | <a href="draft/bug/autogalaxy/power_law_multipole_m1_singular_at_isothermal_slope.md">PowerLawMultipole with m=1 returns -inf / NaN deflections at slope…</a> |
 | 2026-09-27 | filed | <a href="draft/research/autofit/hpc_campaign_epic.md">HPC campaign epic: run large modelling campaigns efficiently…</a> |
 | 2026-09-27 | filed | <a href="draft/feature/autolens_inference/drop_1e3_prior_centring_after_631.md">Drop the 1e-3 prior centring for shear / multipole / ell_comps…</a> |
-| 2026-09-27 | filed | <a href="draft/bug/autoarray/galaxy_duplicate_pytree_registration_after_jax_fitness.md">Building a JAX Fitness then calling register_tracer_classes raises…</a> |
 
 <details><summary>… 10 more (30 left)</summary>
 
 | Date | Event | Task |
 |------|-------|------|
+| 2026-09-27 | filed | <a href="draft/bug/autoarray/galaxy_duplicate_pytree_registration_after_jax_fitness.md">Building a JAX Fitness then calling register_tracer_classes raises…</a> |
 | 2026-09-26 | filed | <a href="draft/feature/pyautohands/bump_colab_urls_autolens_assistant.md">bump_colab_urls.sh: cover autolens_assistant Colab links</a> |
 | 2026-09-26 | filed | <a href="draft/refactor/workspaces/abell_1201_local_cleanup.md">Repo cleanup: retire the Abell 1201 worktree and local raw data</a> |
 | 2026-09-25 | filed | <a href="draft/bug/euclid/latent_total_source_flux_jax_vs_numpy_regression.md">test_latent_euclid_variables_traces_under_jax_jit went red on main…</a> |
@@ -2495,12 +2504,12 @@ The 50 newest things to happen to the work in hand, newest first — issued, par
 | 2026-09-22 | issued | <a href="active/vis_lp_inspection_bundle.md">Support vis_lp-only products in inspection bundles</a> |
 | 2026-09-19 | issued | <a href="parked.md#catalogue-inspection-before-vis-pix">catalogue-inspection-before-vis-pix</a> |
 | 2026-09-18 | parked | <a href="parked.md#fixed-light-numba-s7">fixed-light-numba-s7</a> |
-| 2026-09-18 | filed | <a href="draft/docs/workspaces/model_bullet_parameter_count_drift.md">autolens_workspace: <code>__Model__</code> bullets whose profiles or parameter…</a> |
 
 <details><summary>… 10 more (20 left)</summary>
 
 | Date | Event | Task |
 |------|-------|------|
+| 2026-09-18 | filed | <a href="draft/docs/workspaces/model_bullet_parameter_count_drift.md">autolens_workspace: <code>__Model__</code> bullets whose profiles or parameter…</a> |
 | 2026-09-18 | filed | <a href="draft/bug/pyautomind/arxiv_edge_refuses_the_stdlib_urllib_client.md">arXiv's Fastly edge refuses <code>urllib.request</code> outright — retrying a…</a> |
 | 2026-09-18 | filed | <a href="draft/bug/pyautomemory/arxiv_refs_swallows_every_api_error_silently.md"><code>arxiv_refs.py</code> swallows every API error, so an arXiv outage resolves…</a> |
 | 2026-09-18 | filed | <a href="draft/test/autolens_workspace_test/latent_smoke_assertion_echoes_search_log.md">The latent smoke's empty-block assertion names one cause…</a> |
@@ -2510,12 +2519,12 @@ The 50 newest things to happen to the work in hand, newest first — issued, par
 | 2026-09-17 | filed | <a href="draft/bug/pyautobrain/sizing_faculty_none_rule_keyword_false_judge_triage.md">Sizing faculty: a <code>none</code> rule, keyword hits on prose that only…</a> |
 | 2026-09-17 | filed | <a href="draft/bug/euclid/sed_chain_waveband_fit_can_spin_forever.md">SED chain waveband fit can spin forever in Nautilus exploration…</a> |
 | 2026-09-17 | filed | <a href="draft/maintenance/organs/reduce_session_token_load.md">Reduce the organism's per-session and per-task token load by half</a> |
-| 2026-09-17 | filed | <a href="draft/bug/autofit/nautilus_ep_default_optimiser_reads_use_jax_unguarded.md">Nautilus as an EP default_optimiser crashes on PriorFactors (reads…</a> |
 
 <details><summary>… 10 more (10 left)</summary>
 
 | Date | Event | Task |
 |------|-------|------|
+| 2026-09-17 | filed | <a href="draft/bug/autofit/nautilus_ep_default_optimiser_reads_use_jax_unguarded.md">Nautilus as an EP default_optimiser crashes on PriorFactors (reads…</a> |
 | 2026-09-17 | filed | <a href="draft/refactor/autolens/witt_wynne_solver_library_home.md">Move the Witt–Wynne SIEP solver/projection into PyAutoLens once the…</a> |
 | 2026-09-17 | filed | <a href="draft/bug/autogalaxy/lenscalc_masked_grid_caustic_differs_from_unmasked.md">LensCalc tangential caustic on a masked grid differs ~27 % in…</a> |
 | 2026-09-17 | filed | <a href="draft/refactor/autofit/ep_analysis_level_compile_cache.md">EP: cache the compiled JAX likelihood on the Analysis across factor…</a> |
@@ -2525,7 +2534,6 @@ The 50 newest things to happen to the work in hand, newest first — issued, par
 | 2026-09-16 | filed | <a href="draft/bug/pyautoheart/reusable_smoke_workflow_relevance_gate_skips_custom_runner_tests.md">The reusable smoke workflow's relevance gate skips the <code>unit</code> /…</a> |
 | 2026-09-16 | filed | <a href="draft/feature/autolens/coolest_pixel_grid_export.md">COOLEST: export MGE lens light and pixelized sources as pixel grids</a> |
 | 2026-09-16 | filed | <a href="draft/bug/autoarray/adaptive_mesh_degenerate_pixel_pair.md">Adaptive Delaunay mesh places two mesh pixels 0.001" apart — a…</a> |
-| 2026-09-15 | filed | <a href="draft/maintenance/howtofit/minimum_library_version_stale.md">The workspace staleness warning keys off a value documented never to…</a> |
 
 </details>
 
diff --git a/docs/insight-task-migration.md b/docs/insight-task-migration.md
new file mode 100644
index 000000000..fc5922523
--- /dev/null
+++ b/docs/insight-task-migration.md
@@ -0,0 +1,34 @@
+# Inference task ownership
+
+PyAutoInsight owns inference campaign intent and pending domain tasks. Its
+campaigns.yaml, tasks/, CHECKIN.md and migration.yaml provide one entry point.
+Cortex retains scientific runs, observations and human conclusions. Projects
+retain results and execution; Mind retains bounded implementation PR lifecycle,
+repository claims and mixed library/production epics.
+
+## Migration prepared 2026-10-04
+
+Issue: https://github.com/PyAutoLabs/PyAutoInsight/issues/1
+Source commit: 8dd1edc7e04c960daae3663513c61395823eddb0.
+Five complete prompts and one NUTS/HMC idea fragment are preserved byte-for-byte
+with SHA256 mappings in Insight/migration.yaml. Audit dispositions cover 99
+candidates from 249 current draft/active/registry files.
+
+- prior-centering after Galaxy fix: remains blocked until release verified;
+- Cortex documentation reconciliation: implementation in inference PR18;
+- EP campaign, EP scoping and graphical scoping: preserve phased/decision gates;
+- NUTS/HMC trial on validated likelihoods: needs human decision.
+
+Mixed library implementation epics and scientific run ledgers remain in their
+own repositories; the audit explains each retained candidate. Eight campaign
+rows link authoritative Cortex/project evidence. Historical job ids are not
+proof of jobs running today.
+
+## Landing order and current status
+
+Producer PR18 and Insight destination must merge before source deletion. The
+initial registration PR deliberately leaves all Mind originals intact. Once
+Insight destination merge is verified, run its migration cleanup, regenerate
+Mind registry contents/dashboard, and validate lifecycle/ledger round trips.
+Until then, migration is prepared, not complete. No completed record or claim
+release may assert otherwise. Immutable completion history is not rewritten.
diff --git a/policy/session_start_hook.sh b/policy/session_start_hook.sh
index 54a10fe6e..591c2d7c6 100755
--- a/policy/session_start_hook.sh
+++ b/policy/session_start_hook.sh
@@ -132,12 +132,12 @@ marked_root() {
 holds_an_organ() {
     [ -d "$1/PyAutoBrain" ] || [ -d "$1/PyAutoMind" ] || [ -d "$1/PyAutoCortex" ] \
         || [ -d "$1/PyAutoMemory" ] || [ -d "$1/PyAutoEyes" ] || [ -d "$1/PyAutoHeart" ] \
-        || [ -d "$1/PyAutoHands" ] || [ -d "$1/PyAutoPulse" ] || [ -d "$1/PyAutoNerves" ] \
+        || [ -d "$1/PyAutoHands" ] || [ -d "$1/PyAutoPulse" ] || [ -d "$1/PyAutoInsight" ] || [ -d "$1/PyAutoNerves" ] \
         || [ -d "$1/PyAutoGut" ] \
         || [ -d "$1/organs/PyAutoBrain" ] || [ -d "$1/organs/PyAutoMind" ] \
         || [ -d "$1/organs/PyAutoCortex" ] || [ -d "$1/organs/PyAutoMemory" ] \
         || [ -d "$1/organs/PyAutoEyes" ] || [ -d "$1/organs/PyAutoHeart" ] \
-        || [ -d "$1/organs/PyAutoHands" ] || [ -d "$1/organs/PyAutoPulse" ] \
+        || [ -d "$1/organs/PyAutoHands" ] || [ -d "$1/organs/PyAutoPulse" ] || [ -d "$1/organs/PyAutoInsight" ] \
         || [ -d "$1/organs/PyAutoNerves" ] || [ -d "$1/organs/PyAutoGut" ]
 }
 
diff --git a/repos.yaml b/repos.yaml
index d42b16065..16a7507f5 100644
--- a/repos.yaml
+++ b/repos.yaml
@@ -161,8 +161,15 @@ repos:
     github: PyAutoLabs/PyAutoPulse
     category: organ
     organ: Pulse
-    role: "The Pulse — where the organism feels how fast it runs: the cross-project profiling dashboard over the `<lib>_profiling` project repos (today autolens_profiling) — the instance registry, the versioned `profiling-summary` read contract (v1 live at `autolens_profiling/dashboard/summary.json`), the ingest receipts (resolved commit per project per render) and the Pages board. Validates the exchange contract only; never moves pins, combines unmatched timings, applies the compile threshold to runtime, computes an ecosystem-wide speed score or issues a Heart verdict, and never judges (the Brain's profiling conductor does); the project repos keep their producers, results, drift policy and their own Pages page."
+    role: "The Pulse — where the organism feels how fast it runs: the cross-project profiling dashboard over the `<lib>_profiling` project repos (today autolens_profiling) — campaign intent and pending domain tasks, the instance registry, the versioned `profiling-summary` read contract (v1 live at `autolens_profiling/dashboard/summary.json`), the ingest receipts (resolved commit per project per render) and the Pages board. Validates the exchange contract only; never moves pins, combines unmatched timings, applies the compile threshold to runtime, computes an ecosystem-wide speed score or issues a Heart verdict, and never judges (the Brain's profiling conductor does); the project repos keep their producers, results, drift policy and their own Pages page."
     public_role: "The dashboard of how fast the PyAuto libraries run: it gathers each library's profiling results into one board, so run times and how they change across releases can be seen in one place."
+  PyAutoInsight:
+    path: organs/PyAutoInsight
+    github: PyAutoLabs/PyAutoInsight
+    category: organ
+    organ: Insight
+    role: "Owns inference campaign intent and pending domain tasks, the cross-project inference instance registry, versioned `inference-summary` read contract, ingest receipts and evidence dashboard. Projects own execution, producers and raw samples; Cortex owns scientific run records, observations and human conclusions; Mind owns bounded implementation lifecycle and repository claims. Never infers scientific acceptance from execution, ranks incompatible runs or submits compute on refresh."
+    public_role: "Coordinates inference campaigns and their tasks in one place, showing results, diagnostics, provenance and missing evidence across projects while keeping scientific conclusions with the human and Cortex."
   PyAutoNerves:
     path: organs/PyAutoNerves
     github: PyAutoLabs/PyAutoNerves
diff --git a/scripts/repos_sync.py b/scripts/repos_sync.py
index dcf330487..bae0e760a 100644
--- a/scripts/repos_sync.py
+++ b/scripts/repos_sync.py
@@ -140,7 +140,7 @@ def _repo_resolver(root):
 
 ORGANS = frozenset({"PyAutoBrain", "PyAutoMind", "PyAutoCortex",
                     "PyAutoMemory", "PyAutoEyes", "PyAutoEars", "PyAutoHeart",
-                    "PyAutoHands", "PyAutoPulse", "PyAutoNerves",
+                    "PyAutoHands", "PyAutoPulse", "PyAutoInsight", "PyAutoNerves",
                     "PyAutoGut", "PyAutoScientist"})
 
 
diff --git a/state.json b/state.json
index 0769f4752..1d0c14e35 100644
--- a/state.json
+++ b/state.json
@@ -3,8 +3,8 @@
   "organ": "mind",
   "repo": "PyAutoMind",
   "status": "yellow",
-  "headline": "YELLOW \u2014 24 picks \u00b7 2 in flight",
-  "updated": "2026-10-03T20:07:10Z",
+  "headline": "YELLOW \u2014 24 picks \u00b7 3 in flight",
+  "updated": "2026-10-04T07:53:55Z",
   "pages_url": "https://pyautolabs.github.io/PyAutoMind/",
   "items": [
     {