feat: epics.md registry + dashboard Epics section

AGENTS.md

@@ -44,7 +44,9 @@ For the full workflow narrative, conventions, and registry schemas, read
 - **Registry** — root-level markdown files, each with one job: `active.md`
   (in-flight tasks), `planned.md` (scoped, not started), `parked.md` (started
   but not in flight), `condemned.md` (self-material staged for the Gut's
-  transit-and-void lifecycle — see PyAutoGut), `queue.md` (ordered
+  transit-and-void lifecycle — see PyAutoGut), `epics.md` (long-running
+  multi-phase programmes and the ledger file that holds each one's state),
+  `queue.md` (ordered
   input for `register_and_iterate --queue`), `ideas.md` (raw inbox swept by
   `$intake`, `/intake` in Claude). Mutate these only via the skills in `skills/` so commit
   messages stay consistent.

README.md

@@ -43,7 +43,8 @@ can follow every stage in this repository:
    — the permanent ledger of everything the organism has shipped.
 
 Alongside the lifecycle folders, the root holds the live registry:
-[`active.md`](active.md) (in flight), [`planned.md`](planned.md) (scoped, not
+[`active.md`](active.md) (in flight), [`epics.md`](epics.md) (long-running
+multi-phase programmes), [`planned.md`](planned.md) (scoped, not
 started), [`parked.md`](parked.md) (paused), [`ideas.md`](ideas.md) (raw
 inbox), and [`repos.yaml`](repos.yaml) — the body map naming every repository
 the Mind can direct.

dashboard.html

@@ -59,18 +59,23 @@ <h3>Quick wins <span class="facets">(small enough, and safe enough to run unatte
 <div class="task"><button class="copy" data-cmd="/start_dev draft/maintenance/pyautohands/hands_hygiene_leftovers.md" aria-label="Copy the Claude command">📋</button><p><a href="https://github.com/PyAutoLabs/PyAutoMind/blob/main/draft/maintenance/pyautohands/hands_hygiene_leftovers.md">Hands hygiene: expired announcement dead code, unregistered modules</a> — <span class="facets">pyautohands · small · safe · low</span></p></div>
 <div class="task"><button class="copy" data-cmd="/start_dev draft/maintenance/workspaces/latex_raw_string_docstrings.md" aria-label="Copy the Claude command">📋</button><p><a href="https://github.com/PyAutoLabs/PyAutoMind/blob/main/draft/maintenance/workspaces/latex_raw_string_docstrings.md">Raw-string the LaTeX docstrings emitting SyntaxWarnings (HowToFit + HowToLens)</a> — <span class="facets">workspaces · small · safe · low</span></p></div>
 <div class="task"><button class="copy" data-cmd="/start_dev draft/maintenance/workspaces/script_sizes_snapshot_drift.md" aria-label="Copy the Claude command">📋</button><p><a href="https://github.com/PyAutoLabs/PyAutoMind/blob/main/draft/maintenance/workspaces/script_sizes_snapshot_drift.md">Refresh the stale <code>.script_sizes.json</code> snapshot in @autolens_workspace</a> — <span class="facets">workspaces · small · safe · low</span></p></div>
-<h2>In flight</h2>
+<h2>In flight <a class="facets" 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="/start_dev active/16_transformed_message_factor_gradient_unpack.md" aria-label="Copy the Claude command">📋</button><p><a href="https://github.com/PyAutoLabs/PyAutoMind/blob/main/active/16_transformed_message_factor_gradient_unpack.md"><code>@PyAutoFit</code> <code>TransformedMessage.factor_gradient</code> crashes on first call</a> — <a href="https://github.com/PyAutoLabs/PyAutoFit/issues/1501">issue #1501</a> — <span class="facets">HOLD — do not start dev. Fix-or-delete hangs off the PyAutoFit#1498 logpdf-contract</span></p></div>
 <div class="task"><button class="copy" data-cmd="/start_dev active/jax_default_dependency.md" aria-label="Copy the Claude command">📋</button><p><a href="https://github.com/PyAutoLabs/PyAutoMind/blob/main/active/jax_default_dependency.md">Make JAX a default (non-optional) dependency across the library stack</a> — <a href="https://github.com/PyAutoLabs/PyAutoLens/issues/702">issue #702</a> — <span class="facets">shipped-awaiting-release-followups — ALL ELEVEN PRs merged 2026-08-19 (human-authorized):</span></p></div>
-<h2>Parked</h2>
+<h2>Epics <a class="facets" href="https://github.com/PyAutoLabs/PyAutoMind/blob/main/epics.md">markdown version</a></h2>
+<p class="muted">Long-running multi-phase programmes — 📋 copies a prompt that works out where the epic stands from its ledger and continues it from the next logical point.</p>
+<div class="task"><button class="copy" data-cmd="Continue the 'JAX profiling — inference programme' epic. Its canonical state lives in autolens_profiling/results/notes/inference/PROGRAMME.md — read that ledger (and any DECISIONS/RESULTS files beside it) first. Cross-check this epic's entry in PyAutoMind/epics.md, any related rows in PyAutoMind/active.md, and the referenced repos' open issues and PRs, to work out the last completed phase and what is currently in flight. Then pick the next logical step and continue it through the normal workflow (/start_dev — filing the phase's prompt first if none exists), updating the ledger as the work advances. Note: DECISIONS.md (append-only gate log) and phase_&lt;NN&gt;_*/RESULTS.md sit beside the ledger; slices ship as autolens_profiling issues/PRs, not Mind prompts." aria-label="Copy the Claude command">📋</button><p><b>JAX profiling — inference programme</b> — <span class="facets">ledger: <code>autolens_profiling/results/notes/inference/PROGRAMME.md</code></span></p></div>
+<div class="task"><button class="copy" data-cmd="Continue the 'Cluster strong lensing — Source &amp; Cluster arc' epic. Its canonical state lives in draft/feature/autolens/source_cluster_arc.md — read that ledger (and any DECISIONS/RESULTS files beside it) first. Cross-check this epic's entry in PyAutoMind/epics.md, any related rows in PyAutoMind/active.md, and the referenced repos' open issues and PRs, to work out the last completed phase and what is currently in flight. Then pick the next logical step and continue it through the normal workflow (/start_dev — filing the phase's prompt first if none exists), updating the ledger as the work advances. Note: 12 phased prompts under draft/; issue phases ONE at a time as predecessors near shipping — no bulk issue queues." aria-label="Copy the Claude command">📋</button><p><b>Cluster strong lensing — Source &amp; Cluster arc</b> — <span class="facets">ledger: <code>draft/feature/autolens/source_cluster_arc.md</code></span></p></div>
+<div class="task"><button class="copy" data-cmd="Continue the 'Expectation propagation (EP) campaign' epic. Its canonical state lives in draft/research/graphical_ep/ep_campaign.md — read that ledger (and any DECISIONS/RESULTS files beside it) first. Cross-check this epic's entry in PyAutoMind/epics.md, any related rows in PyAutoMind/active.md, and the referenced repos' open issues and PRs, to work out the last completed phase and what is currently in flight. Then pick the next logical step and continue it through the normal workflow (/start_dev — filing the phase's prompt first if none exists), updating the ledger as the work advances. Note: umbrella phase map — each phase's real content lives in its own prompt under draft/research/graphical_ep/; the campaign file itself is never issued." aria-label="Copy the Claude command">📋</button><p><b>Expectation propagation (EP) campaign</b> — <span class="facets">ledger: <code>draft/research/graphical_ep/ep_campaign.md</code></span></p></div>
+<h2>Parked <a class="facets" href="https://github.com/PyAutoLabs/PyAutoMind/blob/main/parked.md">markdown version</a></h2>
 <details>
 <summary>3 task(s)</summary>
 <div class="task"><button class="copy" data-cmd="/start_dev active/12_single_source_density_refactor.md" aria-label="Copy the Claude command">📋</button><p><b>single-source-density-design</b> — <a href="https://github.com/PyAutoLabs/PyAutoFit/issues/1500">issue #1500</a></p></div>
 <div class="task"><button class="copy" data-cmd="/start_dev active/13_collapse_prior_and_message.md" aria-label="Copy the Claude command">📋</button><p><b>prior-message-collapse-design</b> — <a href="https://github.com/PyAutoLabs/PyAutoFit/issues/1500">issue #1500</a></p></div>
 <div class="task"><button class="copy" data-cmd="/start_dev active/pyautoreduce_slacs1430_acs_comparison.md" aria-label="Copy the Claude command">📋</button><p><b>pyautoreduce-slacs1430-acs-comparison</b></p></div>
 </details>
-<h2>Planned</h2>
+<h2>Planned <a class="facets" href="https://github.com/PyAutoLabs/PyAutoMind/blob/main/planned.md">markdown version</a></h2>
 <details>
 <summary>6 task(s)</summary>
 <div class="task"><button class="copy" data-cmd="/route start the planned PyAutoMind task isothermal-ell-sph-oversampling-at-the-cusp — its record is in planned.md" aria-label="Copy the Claude command">📋</button><p><b>isothermal-ell-sph-oversampling-at-the-cusp</b> — <span class="facets">planned — NOT yet a prompt file; file one via <code>/intake</code> before starting</span></p></div>
@@ -80,7 +85,7 @@ <h2>Planned</h2>
 <div class="task"><button class="copy" data-cmd="/start_dev draft/feature/autogalaxy/piemass_potential.md" aria-label="Copy the Claude command">📋</button><p><b>piemass-potential</b> — <span class="facets">planned</span></p></div>
 <div class="task"><button class="copy" data-cmd="/route start the planned PyAutoMind task latent-nan-guard-honest-run — its record is in planned.md" aria-label="Copy the Claude command">📋</button><p><b>latent-nan-guard-honest-run</b></p></div>
 </details>
-<h2>Backlog</h2>
+<h2>Backlog <a class="facets" href="https://github.com/PyAutoLabs/PyAutoMind/tree/main/draft">markdown version</a></h2>
 <p class="muted">149 filed prompts, not started — sorted most-pickable first (priority, then size).</p>
 <details>
 <summary>bug — 36</summary>

dashboard.md

@@ -183,6 +183,34 @@ Issued — each has an open GitHub issue and usually a branch. The full record f
 
 </details>
 
+## Epics
+
+Long-running multi-phase programmes. Each 📋 prompt has Claude read the epic's ledger, work out where it stands, and continue from the next logical point — no hunting for the paired issue. Full record in [`epics.md`](epics.md).
+
+<details><summary>📋 <b>JAX profiling — inference programme</b> — ledger: `autolens_profiling/results/notes/inference/PROGRAMME.md`</summary>
+
+```
+Continue the 'JAX profiling — inference programme' epic. Its canonical state lives in autolens_profiling/results/notes/inference/PROGRAMME.md — read that ledger (and any DECISIONS/RESULTS files beside it) first. Cross-check this epic's entry in PyAutoMind/epics.md, any related rows in PyAutoMind/active.md, and the referenced repos' open issues and PRs, to work out the last completed phase and what is currently in flight. Then pick the next logical step and continue it through the normal workflow (/start_dev — filing the phase's prompt first if none exists), updating the ledger as the work advances. Note: DECISIONS.md (append-only gate log) and phase_<NN>_*/RESULTS.md sit beside the ledger; slices ship as autolens_profiling issues/PRs, not Mind prompts.
+```
+
+</details>
+
+<details><summary>📋 <b>Cluster strong lensing — Source &amp; Cluster arc</b> — ledger: `draft/feature/autolens/source_cluster_arc.md`</summary>
+
+```
+Continue the 'Cluster strong lensing — Source & Cluster arc' epic. Its canonical state lives in draft/feature/autolens/source_cluster_arc.md — read that ledger (and any DECISIONS/RESULTS files beside it) first. Cross-check this epic's entry in PyAutoMind/epics.md, any related rows in PyAutoMind/active.md, and the referenced repos' open issues and PRs, to work out the last completed phase and what is currently in flight. Then pick the next logical step and continue it through the normal workflow (/start_dev — filing the phase's prompt first if none exists), updating the ledger as the work advances. Note: 12 phased prompts under draft/; issue phases ONE at a time as predecessors near shipping — no bulk issue queues.
+```
+
+</details>
+
+<details><summary>📋 <b>Expectation propagation (EP) campaign</b> — ledger: `draft/research/graphical_ep/ep_campaign.md`</summary>
+
+```
+Continue the 'Expectation propagation (EP) campaign' epic. Its canonical state lives in draft/research/graphical_ep/ep_campaign.md — read that ledger (and any DECISIONS/RESULTS files beside it) first. Cross-check this epic's entry in PyAutoMind/epics.md, any related rows in PyAutoMind/active.md, and the referenced repos' open issues and PRs, to work out the last completed phase and what is currently in flight. Then pick the next logical step and continue it through the normal workflow (/start_dev — filing the phase's prompt first if none exists), updating the ledger as the work advances. Note: umbrella phase map — each phase's real content lives in its own prompt under draft/research/graphical_ep/; the campaign file itself is never issued.
+```
+
+</details>
+
 ## Parked
 
 Started or scoped, not currently in flight — resume by moving the row back to `active.md`. Full detail in [`parked.md`](parked.md).

epics.md

@@ -0,0 +1,28 @@
+# Epics
+
+Long-running multi-phase programmes — work that is bigger than any one task
+and outlives any single issue. Each entry names the epic's canonical
+**ledger**: the file that holds its phase/gate state, wherever it lives. The
+dashboard renders these under In flight with a one-tap resume prompt that
+works out where the epic stands from its ledger and continues it from the
+next logical point — nobody should have to hunt for the issue that pairs
+with an epic's current phase.
+
+Schema per entry: `## <slug>` then `- title:` / `- ledger:` / `- notes:`
+(and optionally `- status:` for a coarse, durable state — never per-phase
+detail, which belongs in the ledger).
+
+## jax-inference-profiling
+- title: JAX profiling — inference programme
+- ledger: autolens_profiling/results/notes/inference/PROGRAMME.md
+- notes: DECISIONS.md (append-only gate log) and phase_<NN>_*/RESULTS.md sit beside the ledger; slices ship as autolens_profiling issues/PRs, not Mind prompts.
+
+## cluster-strong-lensing
+- title: Cluster strong lensing — Source & Cluster arc
+- ledger: draft/feature/autolens/source_cluster_arc.md
+- notes: 12 phased prompts under draft/; issue phases ONE at a time as predecessors near shipping — no bulk issue queues.
+
+## graphical-ep
+- title: Expectation propagation (EP) campaign
+- ledger: draft/research/graphical_ep/ep_campaign.md
+- notes: umbrella phase map — each phase's real content lives in its own prompt under draft/research/graphical_ep/; the campaign file itself is never issued.

scripts/spawn.py

@@ -79,7 +79,7 @@
     # there, and the Mind's root stays the task ledger.
     (".github/AI_POLICY.md", "KEEP_SUB"), (".github/CONTRIBUTING.md", "KEEP_SUB"),
     ("repos.yaml", "SPECIAL:body_map"),
-    ("active.md", "EMPTY"), ("planned.md", "EMPTY"),
+    ("active.md", "EMPTY"), ("planned.md", "EMPTY"), ("epics.md", "EMPTY"),
     ("parked.md", "EMPTY"), ("condemned.md", "EMPTY"), ("ideas.md", "EMPTY"), ("queue.md", "EMPTY"),
     ("autonomy_log.md", "SPECIAL:autonomy_log"),
     # Prompt-file lifecycle (issue #71): draft/ (not-started) -> active/
@@ -193,8 +193,9 @@
 # SHARES a name (e.g. Memory's `bibliography/active.md`, caught by the broad
 # `bibliography/*` EMPTY rule) does not silently inherit a root file's title.
 EMPTY_TITLES = {
-    "dashboard.md": "# PyAutoMind task dashboard",
+    "dashboard.md": "# PyAutoMind Dashboard",
     "active.md": "# Active Tasks",
+    "epics.md": "# Epics",
     "planned.md": "# Planned",
     "parked.md": "# Parked tasks",
     "condemned.md": "# Condemned material",