@bongos/core 1.19.636 → 1.19.638

package/.bongos-core.json

@@ -2,22 +2,22 @@
   "artifact": "bongos-core",
   "manifest_schema": 1,
   "generator": "scripts/gds/package-core.js",
-  "core_version": "1.19.636",
-  "core_contract": "1.19.636",
-  "source_commit": "7e7d5caa1a0c5ff310bdab82a7679fd56e2f71d7",
+  "core_version": "1.19.638",
+  "core_contract": "1.19.638",
+  "source_commit": "723e9b7fc7cb39a1fffb3474441e7d7fb4a9c5b0",
   "source_ref": "HEAD",
-  "built_at": "2026-09-10T03:45:41.270Z",
+  "built_at": "2026-09-10T04:53:40.693Z",
   "redaction": {
     "model": "docs-redacted+functional-verbatim",
-    "docs_redacted": 468,
+    "docs_redacted": 469,
     "agent_docs_stubbed": 24,
-    "functional_verbatim": 2100,
+    "functional_verbatim": 2101,
     "rules": 3,
     "gate_literals": 3,
     "gate": "passed"
   },
-  "file_count": 2592,
-  "tree_sha256": "abf5929f41b5c8b538294d87d01632069ad0371e2e2eed440b97a9727a33d426",
+  "file_count": 2594,
+  "tree_sha256": "337913098795a0534647f0c7d09cb30af3744abd76412a735bb4601433b124c8",
   "files": [
     {
       "path": ".claude/skills/backlog-review/SKILL.md",
@@ -97,7 +97,7 @@
     {
       "path": ".claude/skills/builder-sequence/SKILL.md",
       "mode": "0000644",
-      "sha256": "87a255974f2fe1010bda566f461b08fb68e26aa571ee48e692f5ca9413525a47"
+      "sha256": "7fedadce191a9992c246d86b5853e6d2b196088bd2d944c6166624e240d5df55"
     },
     {
       "path": ".claude/skills/builder-setup/SKILL.md",
@@ -137,12 +137,7 @@
     {
       "path": ".claude/skills/design/SKILL.md",
       "mode": "0000644",
-      "sha256": "f58b813ac63b22f8bbbdc076434866309f6220434412a3aa723af8dff847c38e"
-    },
-    {
-      "path": ".claude/skills/dev/SKILL.md",
-      "mode": "0000644",
-      "sha256": "d0a8ba3e029784a7c8f08a990ae953720fa39b9a86b9d78e1b731bbe7d72e981"
+      "sha256": "6bddf6fda20ec0453af5612a96ff333fbdb140cc1a069a445ce09a439e19720a"
     },
     {
       "path": ".claude/skills/feedback/SKILL.md",
@@ -189,11 +184,6 @@
       "mode": "0000644",
       "sha256": "348bb6e8d27610848fb5f31b672c2c762f56cfb17141f7abe1a8dee0b5030bdd"
     },
-    {
-      "path": ".claude/skills/ideate/SKILL.md",
-      "mode": "0000644",
-      "sha256": "e4279d87f1f4bad8566bb69ea82278bde3a725a3879f972ad3cb34bb2cef9de5"
-    },
     {
       "path": ".claude/skills/merge-mode/SKILL.md",
       "mode": "0000644",
@@ -229,11 +219,6 @@
       "mode": "0000644",
       "sha256": "53bfcf6c531c0aeefc1696b470ba6a746f4490501ab092766f42c501f06d453d"
     },
-    {
-      "path": ".claude/skills/paint/SKILL.md",
-      "mode": "0000644",
-      "sha256": "05a42301ac128a9a7abe3aa93158a9005af5b45afed623e8bc8cb92380d09fb3"
-    },
     {
       "path": ".claude/skills/planning-session/SKILL.md",
       "mode": "0000644",
@@ -1889,10 +1874,15 @@
       "mode": "0000644",
       "sha256": "2fd3191d95758b5be0f84dd050cfa0244707c0c86646eaa9dd843216e754ff07"
     },
+    {
+      "path": "docs/adr/0274-one-kernel-three-role-packs.md",
+      "mode": "0000644",
+      "sha256": "0dad96decb62fe911ab644dad16e70a0ffe2482dfe12bcec9f2dd6329733b97d"
+    },
     {
       "path": "docs/adr/README.md",
       "mode": "0000644",
-      "sha256": "eeb226dcbb35f522095d294952379ea8647df9f8f5048eb1754eb112e9b687af"
+      "sha256": "c62aaf29b37052be529f9a9d1a844af63d59b8bd864f49df816443d3ed1fd679"
     },
     {
       "path": "docs/api-reference.md",
@@ -2767,7 +2757,7 @@
     {
       "path": "docs/file-map.md",
       "mode": "0000644",
-      "sha256": "0cec3b144d8cca893b49c1d08bdacf37220dc1b670fd771656dda659bae16edc"
+      "sha256": "acb9e6949bcdd928e6c9b09d6c0ab0d45a065bab8f97e323b9f7554138bece47"
     },
     {
       "path": "docs/handoff-template.md",
@@ -2777,7 +2767,7 @@
     {
       "path": "docs/module-api-changelog.md",
       "mode": "0000644",
-      "sha256": "74c46d5efae11f3adc3f0cb79bec6757e0f7bb820e7f77c1f66cd6fc083a503c"
+      "sha256": "d6ec97aa92f3776d80dd7f0e78b3f4428b356fc7ec327b0132514ab48564dd56"
     },
     {
       "path": "docs/modules-contract.md",
@@ -2847,7 +2837,22 @@
     {
       "path": "docs/onboarding/slash-commands.md",
       "mode": "0000644",
-      "sha256": "8e8d5c7d0843faa2dc550a640676a824c9be319c751cf1a1cb19eaebeb86b153"
+      "sha256": "174c90095ed45bf107852c017493fdd0fa3f07bdf8fbe69c09ebf77bc00747f2"
+    },
+    {
+      "path": "docs/packs/artist.md",
+      "mode": "0000644",
+      "sha256": "27d49ea30d9efb2d314f626df46edcc1f2d86e3a08b0a33581d78dde4b42708f"
+    },
+    {
+      "path": "docs/packs/engineer.md",
+      "mode": "0000644",
+      "sha256": "a3f50731d3e4d536e6b7e98ed533f444e2efc757e914ab8ad8bff0d1c68a2509"
+    },
+    {
+      "path": "docs/packs/ideator.md",
+      "mode": "0000644",
+      "sha256": "76404a9e4e5e4b47942c1701f62c811f23fdf6f67d274e47318185bd571c03ea"
     },
     {
       "path": "docs/project-context.template.md",
@@ -5912,7 +5917,7 @@
     {
       "path": "modules/lifecycle/ship-card.js",
       "mode": "0000644",
-      "sha256": "2b6121aacc015dd7edd4eb1acb3ca9cf5f1a0adf62cd6dbe3f92d342717f88a4"
+      "sha256": "c0e5a079c586e689d569d764526be5d99808262c9773ef17902e2416047494ab"
     },
     {
       "path": "modules/lifecycle/ship-preflight.js",
@@ -7707,12 +7712,12 @@
     {
       "path": "package-lock.json",
       "mode": "0000644",
-      "sha256": "7e8bd5c903c8f50fa7bcc80c7e317a4ea202bf079e0e63a8beafc988718953e1"
+      "sha256": "6937e64c5b6736b161c21f2dda42f385a1e3ed8499ea9bb39761c99bfcc3038c"
     },
     {
       "path": "package.json",
       "mode": "0000644",
-      "sha256": "0e2c886ce78f829a2b694c3c49af7fafff6297695da0ddb346d7c0936fe95ff8"
+      "sha256": "a4cb1f8d57a319db2ddce580a2267e7d1d32ba37a3d87f75ef5fd03ed3271837"
     },
     {
       "path": "public-docs/index.html",
@@ -7957,7 +7962,7 @@
     {
       "path": "scripts/gds/claim.js",
       "mode": "0000644",
-      "sha256": "d925dfa3831cb3bd16b2ff1761344bd4bedf7d916613fffd9dc5d9574244f525"
+      "sha256": "07603f4750d2be31e1399d61df6d1bad0dbcd62ef4784077c9030b7c091d648e"
     },
     {
       "path": "scripts/gds/classify-discipline.js",
@@ -8087,7 +8092,7 @@
     {
       "path": "scripts/gds/discipline-modes.json",
       "mode": "0000644",
-      "sha256": "c0c1849145999ef02f3022b22e718da8bb4fb680020124d115423f2f868834d1"
+      "sha256": "a62a714209198f02ca4f964ecfd1211dd3712b946f049843a7ec921b5907f1d0"
     },
     {
       "path": "scripts/gds/discord-channels.js",
@@ -8172,7 +8177,7 @@
     {
       "path": "scripts/gds/fitness.js",
       "mode": "0000644",
-      "sha256": "610d76cee575fa1a87f2ff7bb84f5576f3c4a8cf2e727e6c372d76ebf9a0a340"
+      "sha256": "4d1f5959f832702b28a5c514acec179a605c4f4d7be5d13c2d4acb3f094f7ba0"
     },
     {
       "path": "scripts/gds/gate-review.js",
@@ -8372,7 +8377,7 @@
     {
       "path": "scripts/gds/main-audit.js",
       "mode": "0000644",
-      "sha256": "18e14d00fead3293eb5b1c9d7a756a0a3780147938a40253ff35540740a73f4b"
+      "sha256": "35047f05c6059aee0cb64b13ef1495e4db1f3885b5b9c8cac5c60283069a059d"
     },
     {
       "path": "scripts/gds/me.js",
@@ -8527,7 +8532,7 @@
     {
       "path": "scripts/gds/publish-manifest.js",
       "mode": "0000644",
-      "sha256": "08c85bcb004f5a1b408a1cefcb7e47757ea42f364a29629793bd0b76771e46d2"
+      "sha256": "722fba8fbec20b32627304defc05533d9f4f8688b2f094ed554e82dfa9dabe1e"
     },
     {
       "path": "scripts/gds/pulse.js",
@@ -8604,6 +8609,11 @@
       "mode": "0000644",
       "sha256": "084fa5c4da1e755fa0edde5ec770e5151a9f66a247bca222bc7a2366ccf61748"
     },
+    {
+      "path": "scripts/gds/role-pack-guard.js",
+      "mode": "0000644",
+      "sha256": "a2203bccbe0c91882ad355e86874de5584343064321f40a1e72c89a760545c74"
+    },
     {
       "path": "scripts/gds/route-shadow-guard.js",
       "mode": "0000644",
@@ -9457,7 +9467,7 @@
     {
       "path": "src/module-api.js",
       "mode": "0000644",
-      "sha256": "4c4fcf6a9833c72d6c371208d00afc60eba49cd07e5a0c9b4cd384133089340d"
+      "sha256": "b988076c42210125216247e5b677c2351ba4e668c9bfac9e3c95028227f97108"
     },
     {
       "path": "src/module-loader/catalog.js",
@@ -10422,7 +10432,7 @@
     {
       "path": "tests/edit_hints.mjs",
       "mode": "0000644",
-      "sha256": "1139ea32e7f723d97ac63dd2c862896945b69970ac5edbe0368fb427ff0fe3b2"
+      "sha256": "bf15f163f4c07063c10f477168f9f45a2216070d56e7720ba3e60e17a95f221d"
     },
     {
       "path": "tests/effective_visibility_predicate.mjs",
@@ -10487,7 +10497,7 @@
     {
       "path": "tests/fitness.mjs",
       "mode": "0000644",
-      "sha256": "2208a1e4d3df3ce712c4e4dbc7908d20520833375a2c07c3da5d37252dd56b7a"
+      "sha256": "163a9dd1851a44c38acb1210fbd7078561f9f8fd5fc54cae92734383384e83a7"
     },
     {
       "path": "tests/fitness_gds_ratchet.mjs",
@@ -10857,7 +10867,7 @@
     {
       "path": "tests/government_protected_surfaces.mjs",
       "mode": "0000644",
-      "sha256": "fd05f72404be0d007222793a6ed3a49904239f8bb354cddab6b052d74288b3b1"
+      "sha256": "cd56facc3522f17cd83b067ee1ae9dcb3e4e03dac13157013c44d7ba23508af2"
     },
     {
       "path": "tests/government_rank_sync.mjs",
@@ -11202,7 +11212,7 @@
     {
       "path": "tests/idea_ideate_full.mjs",
       "mode": "0000644",
-      "sha256": "5000ed6a7b1a24eced0f070e556af354c8fa9c7937f43c7b932a3ca29f8a5f35"
+      "sha256": "af36a9663bca5653d525ded5adc73a78bf507bc5400672e3ee69985e93f37174"
     },
     {
       "path": "tests/idea_quick_guard.mjs",
@@ -11442,7 +11452,7 @@
     {
       "path": "tests/main_audit.mjs",
       "mode": "0000644",
-      "sha256": "9a1813d6036886ea07d1563e71a466e66c28554b75b22605464c795fd9fa2ebb"
+      "sha256": "eff2ce96889306f4898686c3648d302f54fff5f137ee41ddf543866fdf2f807d"
     },
     {
       "path": "tests/maintenance_goal.mjs",
@@ -11557,7 +11567,7 @@
     {
       "path": "tests/module-scope-map.mjs",
       "mode": "0000644",
-      "sha256": "470ccf8f8d5f30ce613f01db7dca4ceda00163b5df16a8a0a4ec84a947eba083"
+      "sha256": "bd6fa3e2a2045af4d62a3097aeb38a5e191c7e954de4b0486de035bb3dc032ba"
     },
     {
       "path": "tests/module_api.mjs",
@@ -12072,7 +12082,7 @@
     {
       "path": "tests/publish_manifest.mjs",
       "mode": "0000644",
-      "sha256": "5eca5f50f200166ce0fd7e40a57e18fffb135139fe558d0dda34064d426814dc"
+      "sha256": "b6f9f357277902082dde8578531059e59ae9fdbd9c2194ea72fb0a71405d5ea2"
     },
     {
       "path": "tests/publish_reconciler.mjs",
@@ -12482,7 +12492,7 @@
     {
       "path": "tests/ship_card.mjs",
       "mode": "0000644",
-      "sha256": "692c96f61a6d75568c927aa754707c5860effe63fe31c476e3eb82b6d066c9b6"
+      "sha256": "c55dddec249ec936a1278f1d48edea1adf15003cc454a8bfb9b1b891160be865"
     },
     {
       "path": "tests/ship_card_injection.mjs",

package/.claude/skills/builder-sequence/SKILL.md

@@ -6,7 +6,7 @@ description: >-
 
 You are running a **chain**: several dependency-linked tasks, one after another, from a single kickoff. Two-thirds of the open backlog sits behind a predecessor, and shipping a task auto-promotes its dependents `backlog`→`ready` — but nothing picks the newly-unblocked task up. That gap is what this skill closes.
 
-This skill owns the **loop and the halt discipline**. It does not re-derive how to claim, build, or ship — `/builder-claim`, `/dev`, and `/builder-ship` own those, and you defer to them each pass.
+This skill owns the **loop and the halt discipline**. It does not re-derive how to claim, build, or ship — `/builder-claim`, the claimed task's role pack (docs/packs/), and `/builder-ship` own those, and you defer to them each pass.
 
 ## The one rule
 
@@ -32,7 +32,7 @@ This skill owns the **loop and the halt discipline**. It does not re-derive how
 
 3. **Claim it** — `/builder-claim <id>`, or `node scripts/gds/claim.js <id>`.
 
-4. **Build it** — follow `/dev`. Stay strictly inside the task's `touches[]`. Everything the normal task loop requires still applies: read the acceptance criteria, work test-first, commit before shipping.
+4. **Build it** — follow the claimed task's role pack (`docs/packs/engineer.md` for engineer work). Stay strictly inside the task's `touches[]`. Everything the normal task loop requires still applies: read the acceptance criteria, work test-first, commit before shipping.
 
 5. **Ship it** — `/builder-ship`. The task must reach `shipped`, not stall at `completed` or `confirmed`. A ship that does not land has not unblocked anything, so the chain cannot advance; treat it as a halt (see `ship_failed` below).
 

package/.claude/skills/design/SKILL.md

@@ -4,7 +4,7 @@ description: >-
   The UI design session — the operating playbook for ui-discipline work (layouts, pages, components, design systems, the hall / status / landing surfaces) on any Cloud Bongos instance. Show-first like /paint and world-first: the branding pack and DESIGN.md are read before any taste rule. Triggers: "/design", "design this page", "let's work on the UI", or a claimed ui task routing here.
 ---
 
-You are running a **UI design session**. Two things make it different from `/dev`: the work is judged **by looking at it** (so every step ends in a rendered screen, not a paragraph), and it is done **inside a world that is not yours** — the instance's, which you read first and never override. The `ui-design` module's [CLAUDE.md](../../../modules/ui-design/CLAUDE.md) is the reference; this file is the loop.
+You are running a **UI design session**. Two things make it different from the Engineer pack (`docs/packs/engineer.md`): the work is judged **by looking at it** (so every step ends in a rendered screen, not a paragraph), and it is done **inside a world that is not yours** — the instance's, which you read first and never override. The `ui-design` module's [CLAUDE.md](../../../modules/ui-design/CLAUDE.md) is the reference; this file is the loop.
 
 ## Step 0 — the tools, then the mode
 
@@ -103,4 +103,4 @@ A look that is not listed here is not in the library. Adding one is the README's
 ## How sessions reach this skill
 
 - **Standalone:** invoke `/design` to have the loop in front of you.
-- **From a claim:** a `ui`-discipline task makes `claim.js` print the `/design` directive from `scripts/gds/discipline-modes.json` — one carrier, one lookup, the same map `/dev`, `/paint` and `/ideate` ride.
+- **From a claim:** a `ui`-discipline task makes `claim.js` print the `/design` directive from `scripts/gds/discipline-modes.json` — one carrier, one lookup, the same registry the three core crafts ride to their role packs under `docs/packs/` (task 1002990).

package/docs/adr/0274-one-kernel-three-role-packs.md

@@ -0,0 +1,55 @@
+# ADR 0274 — One kernel, three role packs: the always-loaded instructions stop being an engineer's
+
+**Date:** 2026-09-10
+**Context:** BONGOS-V2, goal [1000095](https://cloudbongos.com/builders#/goal/1000095) (Working area 6 — Governor / Builder / Artist / Ideator experience), criterion `wa6-kernel-and-packs`, task [1002990](https://cloudbongos.com/builders#/task/1002990). Builds on [ADR 0061](<redacted>.md) (the CLAUDE.md context budget), [ADR 0062](<redacted>.md) §7 (the core ↔ host split), [ADR 0272](<redacted>.md) (a craft is an identity, not a toolchain) and [ADR 0271](<redacted>.md) (which predicted this move: *"when declaration lands the contract MOVES into the Ideator pack"*). Its dependency, task 1002989, widened the module map to cover the methodology surfaces so this work had a wall to stand in.
+**Status:** Accepted.
+
+## Problem
+
+The root `CLAUDE.md` loads **in full, every session, for every craft** — and it was an engineer's document. 250 lines and 32,422 characters of claim-before-working, worktree-per-claim, `DEPS_NOT_SHIPPED`, doc-comments, `touches[]`, generated-artifact hygiene, `/merge-mode`. An ideator who came to think about an idea, and an artist who came to look at something, each paid the full token cost of a build pipeline they were not going to run, and read it as instructions addressed to them.
+
+That is a correctness problem before it is a budget one. The file said, as its hard rule, *"every change is backed by a claimed Bongos task, no exceptions"* — while the same goal's criterion `wa6-drift-stops-the-session` says an ideator is *"pointed at ideation (which needs no claim)."* The kernel and the criterion disagreed, and the kernel is what every session actually read.
+
+Meanwhile the craft-specific material that *did* exist lived in three slash-command skills — `/dev`, `/ideate`, `/paint`. A skill is the wrong carrier for a role's operating playbook for two reasons. Its `description` is resident in every session's skill listing whatever craft is running (the listing measured 27,022 chars against a ~8,000 budget, and `dev`/`ideate`/`paint` were three of the twenty over-length entries). And it must be *invoked* — a session that never types the slash command works from the kernel alone, which is the engineer's document again.
+
+## Decision
+
+**One role-neutral kernel plus one pack per craft, bound by a registry that CI checks.**
+
+1. **`CLAUDE.md` is the kernel.** What Cloud Bongos is, the lifecycle in plain terms, the money, the trust boundary, the identity split, the content charter, and the navigation index. The test for a line is now two-part: *would removing it cause a mistake?* **and** *is it true for every craft?* A line that is only true for engineers moves out. Result: 250 → 216 lines, 32,422 → 22,572 chars, which clears the `fitness.js` character target it had been over.
+
+2. **Three packs under `docs/packs/`** — `engineer.md`, `artist.md`, `ideator.md`. **On the name:** task 1002990's own title says *"the Builder pack"*, and this ADR does not. The criterion it implements names the three crafts *Engineer, Artist, Ideator*; `CORE_DISCIPLINES` is `['engineer','artist','ideator']` (ADR 0272); and `tasks.discipline` stores `engineer`. "Builder" is the person who holds a claim in any craft — an artist is a builder too — so naming one craft's pack after it would collide with the word the rank ladder and the whole hall already use. The pack is keyed by the discipline it is bound to, because that key is what the registry looks up. The Engineer pack inherits today's methodology verbatim in substance (the loop, the working rules, the session protocol, the daily cadence, subagent routing) plus what `/dev` carried. The Ideator pack absorbs `/ideate` and writes down the two grades the criterion names: a quick idea stays **one line**, and after one is filed Claude **asks** whether to develop it into a Full Idea — five questions drawn out of the conversation, in the ideator's own words, scored on completeness. The Artist pack absorbs `/paint` and generalises it off the pixel-art pipeline (ADR 0272: the craft is core, the pipeline is an optional host module): show-first, **as little interpretation as possible** — take the brief literally, you are their hands and not their art director — and **arrive precalculated**, with the look already loaded and the options already rendered, because a session that opens with intake questions is a form.
+
+3. **`/dev`, `/ideate` and `/paint` are deleted.** The skill listing drops to 25,624 chars and 17 over-length descriptions.
+
+4. **`scripts/gds/discipline-modes.json` becomes the role registry.** A core craft names a `pack` (a file to *read*); a module-contributed discipline may still name a `skill` (a command to *invoke*), which is how `ui` → `/design` survives unchanged, as ADR 0272 requires. `claim.js` prints whichever the entry carries, with the right verb.
+
+5. **`scripts/gds/role-pack-guard.js` hard-fails CI** when an entry names a pack that is not in the tree, when an entry names neither pointer, when fewer than three crafts carry a pack, or when a pack exceeds its character budget (16,000 hard / 13,000 target, ratcheted just above the packs as authored).
+
+### Why the guard is hard when the runtime is fail-open
+
+`claim.js` must never break a granted claim, so its registry read swallows everything. That means a pack path pointing at nothing degrades to **silence**: the builder is told their craft's label and nothing else, and the playbook that carries everything the kernel deliberately dropped never loads. The role is hollow and no error says so. CI is therefore the only place that failure can be caught — which is exactly what the criterion asks for when it says *"CI budgets each pack."*
+
+### Why `docs/packs/` and not `.claude/packs/`
+
+Three reasons, in order of weight. `docs/` is already owned by the `methodology` key in `module-scope-map.js`, so the packs sit inside a wall that exists; a new top-level path would resolve to **no** key, and the scope map's own comment records what that costs — a missing key is an empty wall, and `goal-scope-check` then denies every member-authored task (the 1002690 / 1002736 / 1003124 precedent). Second, a pack must not be a skill: skills are listed in every session regardless of craft, which is the cost this ADR is removing. Third, `docs/` resolves from the core package the same way `discipline-modes.json` does, so a scaffolded instance reads the registry and the pack it names from one place with no materialisation step.
+
+`docs/` is default-deny in the publish manifest, so `docs/packs/` is registered explicitly. The registry line and the allowlist line are only ever added or removed together — a core that shipped the registry without the packs would print, on every claim on every instance, a path that resolves to nothing: precisely the silent failure the guard exists to make loud.
+
+## Consequences
+
+- **The ledger rule survives the split, and gets more honest.** The kernel keeps it in role-neutral form — *a change to project state is recorded in Bongos before it is made* — and each pack says what recording means for that craft. The engineer holds a claim; the ideator files the idea; the artist claims the work or records the review. The no-loopholes paragraph moves to the Engineer pack intact, where it is true.
+- **The ship card had to move with the claim.** `modules/lifecycle/ship-card.js` routed a claimed task's "Open the work" button to `/dev`, `/paint` or `/ideate`. Left alone it would have shipped a button firing a command that no longer exists — the same divergence task 1003487 fixed in the other direction. It now points at the pack, and `tests/ship_card.mjs` asserts no core craft is routed to a slash command.
+- **The ship card duplicates the registry, and a test is the shared source it cannot have.** `ship-card.js` is a module and may not require a core script across the boundary, which is why it and `claim.js` have always carried the map twice — and task 1003487 was exactly that duplication drifting. Rather than pretend the duplication away, `tests/fitness.mjs` now walks the registry entry for entry and asserts the card agrees, so a pack rename fails a test instead of shipping a hall button that opens nothing.
+- **`tests/idea_ideate_full.mjs` moved with its subject.** Its thirteen assertions pinned the Full-Idea contract in `/ideate`'s prose — the same-route rule, the never-invent rule and its reason, the five-questions-are-not-restated rule. Repointing it at `docs/packs/ideator.md` is what stops the split from deleting a guard along with the file it guarded; all thirteen pass against the pack unchanged.
+- **The line target is not yet met.** The kernel is 216 lines against a ~200 target — a `fitness.js` warning, not a failure, and the same warning it carried before at 250. The character target, which is the real per-session cost, is now cleared. Tightening the line count is the context-budget cluster's work (tasks 1003620, 1003687, 1003688), which the goal's audit sequenced explicitly behind this one.
+- **Governor is deliberately absent.** Three packs, not four, until the owner decides. The guard's floor is `>= 3`, so adding a fourth is additive and removing a third is a failure.
+- **What this unblocks.** Task 1002991 (serve composed session context from the server) and 1002992 (inject role context at start and claim) now have something to compose and inject: today the pointer is printed and read, and those tasks replace the reading with injection. Task 1003732 authors the three responsibility statements into the packs' reserved seams — the Artist pack names its own, unwritten, rather than inventing it.
+
+## Rejected
+
+- **Keeping the three skills and adding packs beside them.** Two homes for one craft's playbook is how they drift, and it keeps the skill-listing cost the split exists to remove.
+- **Packs as skills under `.claude/skills/packs/`.** Every session pays for every pack's description — the exact failure being fixed.
+- **A fourth Governor pack "while we're in here."** The criterion defers it by name, and inventing a role's operating playbook without its owner is how the un-owned areas in this goal's audit came about.
+- **Duplicating the safety rules into all three packs.** The charter forbids duplication for a reason: three copies drift and the kernel becomes the one nobody trusts. Safety stays in the kernel; craft stays in the pack.
+- **Leaving `checkRolePacks` advisory, like the doc-comment check.** That check is advisory because a blocking doc-comment gate buys filler prose. This one gates a *resolvable path*, which is a fact and not a judgement, and the runtime cannot report it.

package/docs/adr/README.md

@@ -365,3 +365,4 @@ This keeps the decision history honest and traceable.
 | 0271 | [**The ideator surface is two halves: Bongos owns the prose, the harness owns the chrome** ([task 1003506](https://cloudbongos.com/builders#/task/1003506) · goal 1000095 — *Working area 6, Governor / Builder / Artist / Ideator experience*, criterion `wa6-kernel-and-packs`; owner idea 1000732, 2026-08-16). The ask — *no need for an ideator to see all the text and technical work, or the inner Bongos logic* — is two problems in one sentence. **The PROSE half ships here**, through three seams that already existed: a `machinery` field on `interaction-prefs` (per-PERSON, default `show`, riding the live Conductor injection rail — advisory 8, which means task [#1002514](https://cloudbongos.com/builders#/task/1002514)’s claim that `contractFor()` has zero consumers is STALE); the `ideator` directive in `discipline-modes.json` (per-ROLE, the file whose own docs say *to give a discipline its own experience later, add an entry here*); and a new surface section in `/ideate`, which the directive names as the only home of the experience. Default `show` means no existing builder changes. **The CHROME half is the harness’s and is deferred to a spike, not declared impossible** — this record was nearly written on the wrong assumption that config cannot reach it. Two real levers exist: **A**, shipped client defaults (`/focus` renders *only your last prompt, a one-line tool-call summary, and the final response*; `alwaysThinkingEnabled:false` stops thinking being generated; `showTurnDuration`/`spinnerTipsEnabled`/`statusLine` trim chrome) — but `/focus` is fullscreen-only (research preview), `viewMode`’s committability is UNVERIFIED, and thinking suppression is per-model; and **B**, an app-owned surface over `claude -p --output-format stream-json`, whose documented streaming-UI pattern (working indicator on a `tool_use` block, render on `result`) IS the ask, is model-independent, and is the recorded bias. **Bypass is REJECTED** despite the idea proposing it, on three grounds ending in a technical one: the least technical role is the worst audience for removing confirmations; the prompt was never the gate ([ADR 0016](<redacted>.md) — rank is checked server-side, and sub-Metic protected-path globs already narrow an ideator); and a project **cannot** ship it — `defaultMode: bypassPermissions` in `.claude/settings.json` silently does not take effect and starts the session in Manual. Rejected also: a hall ROOM on the Studio’s precedent (goal 1000074’s *a place, not a form*) — right for an artist, wrong here, because the ideator’s complaint is about the Claude session’s surface, which the hall does not render; and waiting on per-session declared role, whose dependency is softer than filed since `discipline` + the claim-time fork already carry the craft axis. When declaration lands the contract MOVES into the Ideator pack with nothing undone.](<redacted>.md) | roles / ideator experience / session surface |
 | 0273 | [**The hall nav can gate on a permission, and the Board Room leaves Government** ([task 1003744](https://cloudbongos.com/builders#/task/1003744) · goal 1000111 — *Working area 7, Government*, criterion `wa7-government`; owner decision 2026-09-09). `shell.js` `gateAllows()` resolved six gate keys and every one was a RANK, so a nav item whose page is gated on an ATOM had to borrow the nearest rank floor and hope the two agreed. [ADR 0266](<redacted>.md) ended that: the Board Room page asks `board.vote.cast`, which floors at XENOS deliberately (`board.js` — the floor is low *so that widening the board works*), while its nav item read the Government group's metic floor; the carve filed the gap rather than guessing at it. Latent today because the constitution seats `rank:archon`, live the moment an instance widens membership below Metic — the exact act [ADR 0175](<redacted>.md) §6 exists to make a CONFIG change — at which point those members reach the room by URL and by the `board_votes` deep link with no sidebar entry. **Decision: a `perm:<atom>` gate key, answered by the SERVER.** `NAV_PERMISSION_ATOMS` in `routes/me.js` is a frozen allowlist of the atoms a nav item may gate on; `/me` resolves each through the `government` port and returns `permissions: { atom: boolean }`; `gateAllows` reads that map and never re-derives it (the social-mode precedent — a page that decided for itself could disagree with the rest of the hall); the Board Room item declares the SAME key `requireBoardVotePage` checks, so the two cannot drift. An ALLOWLIST rather than the full resolved set, because the full set is the shape of the instance's governance and a cosmetic sidebar has no business shipping it on every page load. `permissions: null` (government off, port missing, resolver threw) means *cannot say*, not *holds nothing*, and HIDES the link — `default: false` and OFF-is-the-fallback, applied. Cosmetic throughout ([ADR 0016](<redacted>.md)): the page gate and each route's `requirePermission` are the wall. **The placement was the owner's call, and the reason this task existed:** gating honestly while leaving the item in the Government group would put a WIDER item under a narrower head, so a newcomer board member would see a group labelled "Government" holding one link — more authority than they hold. Declined; the item MOVES to the unlabeled project group after Tasks, because a sitting is work the project does, not a monitoring surface. Government's audience, head and floor are unchanged and it is rank surface throughout again. Also rejected: widening the Government head to show whenever any item inside is visible, and a single-item "Board" group. `tests/hall_nav.mjs` gained the MISSING HALF of its group invariant — it checked only that a head is never wider than its items, never that an item is wider than its own head (an orphan link under a hidden label, exactly the declined arrangement) — and `GATE_WIDTH` learned `government`, which had been `undefined` so every comparison against that group passed vacuously. The harness `/me` fixture carries `permissions`, or the item would be permanently invisible in `scripts/hall-preview`. **Not done:** no other item re-gated — `people` still reads `government` for `project.curate`.](<redacted>.md) | hall nav / permissions / government |
 | 0272 | [**A craft is an identity, not a toolchain: `artist` becomes core** ([task 1003719](https://cloudbongos.com/builders#/task/1003719) · goal 1000111 — *Government*). Reported from Discord on 2026-08-15 as "cant save preferences of role types in account page"; reproduced live 2026-09-09: `GET /me` returns `preferred_disciplines: ["artist","ideator"]` and `PATCH /me/disciplines` with that SAME array returns `422 invalid_discipline {bad:["artist"]}` — the API handing a builder a value and refusing to accept it back. **Cause:** `updateBuilderDisciplines` validates against `enabledDisciplines()`, and R59 (task 1198) put `artist` in the module half on the reasoning that the craft existed to make the game pixel art, so it rode `art-pipeline`. **That gate had stopped existing** — art-pipeline was never converted to a discovered module, there is no `modules/art-pipeline/module.json`, and so NOTHING declared `contributes.disciplines: ["artist"]`. Measured on a FULL repo checkout with the pipeline present on disk: `["engineer","ideator","ui"]`. The craft was offered by no instance in any configuration, and is unreachable by construction on a published core anyway, since [ADR 0098](<redacted>.md) excludes `modules/art-pipeline/` for a hard copyright reason (ripped reference maps). Meanwhile it stayed first-class everywhere else: the `builders.preferred_disciplines` CHECK (086/176), `tasks.discipline`, and the help-request craft roster all accept it — so a TASK could be an artist task and a PERSON could not be an artist. **The tests were green because they skipped:** the artist assertion sat behind `if (!EFFECTIVE["art-pipeline"]) return;`, always false, so the only real leg never ran. **Decision:** `artist` joins `CORE_DISCIPLINES` → `["engineer","artist","ideator"]` (migration 176 enum order); modules still contribute their own (ui-design → ui). A craft names WHO A BUILDER IS, not which toolchain ships — an instance can have artists without a pixel-art pipeline, and binding the identity to one optional module was the category error. The new invariant, pinned with no skip guard over the widest module set: anything the DB CHECK will store, some instance shape must offer. Rejected: enabling art-pipeline (impossible — absent from the package, and enabling an absent module throws at boot); writing the missing manifest (works only in a full checkout); an instance-declared discipline list (leaves the core CHECK describing a craft core cannot offer, and does not fix the vanilla case); stripping unofferable crafts from stored rows (deletes a builder stated identity to make a validator mistake look consistent). No migration — the rows were always CHECK-valid; only the validator disagreed.](<redacted>.md) | crafts / instance composition |
+| 0274 | [**One kernel, three role packs: the always-loaded instructions stop being an engineer's** ([task 1002990](https://cloudbongos.com/builders#/task/1002990) · goal 1000095 — *Working area 6, Governor / Builder / Artist / Ideator experience*, criterion `wa6-kernel-and-packs`). The root `CLAUDE.md` loads in full every session for every craft, and it was an engineer's document: 250 lines / 32,422 chars of claim-before-working, worktree-per-claim, `DEPS_NOT_SHIPPED`, `touches[]` and `/merge-mode`. An ideator and an artist each paid the full token cost of a pipeline they were not going to run and read it as instructions addressed to them — and it CONTRADICTED the same goal's `wa6-drift-stops-the-session`, which says ideation *"needs no claim"*, while the kernel's hard rule said every change is backed by one. The craft-specific half lived in three SKILLS (`/dev`, `/ideate`, `/paint`), the wrong carrier twice over: a skill's description is resident in every session's listing whatever craft is running (27,022 chars against a ~8,000 budget, three of them over-length), and a skill must be INVOKED — a session that never types the command works from the kernel alone, which is the engineer's document again. **Decision: a role-neutral kernel plus one pack per craft under `docs/packs/`, bound by a registry CI checks.** The kernel keeps what is true for every craft (the lifecycle, the money, the trust boundary, the identity split, the charter, navigation) and drops to 216 lines / 22,572 chars, clearing the character target it had been over; the per-line test gains a second half — *is it true for every craft?* The Engineer pack inherits today's methodology; the Ideator pack absorbs `/ideate` and writes down the two grades the criterion names (a quick idea stays ONE LINE, then Claude ASKS whether to develop it into a Full Idea — five questions drawn out of the conversation, in the ideator's own words, scored on completeness); the Artist pack absorbs `/paint`, generalised off the pixel-art pipeline per [ADR 0272](<redacted>.md) — show-first, as little interpretation as possible, and ARRIVE PRECALCULATED, because a session that opens with intake questions is a form. The three skills are deleted. `scripts/gds/discipline-modes.json` becomes the ROLE REGISTRY: a core craft names a `pack` (a file to read), a module-contributed discipline may still name a `skill` (`ui` → `/design`, unchanged as ADR 0272 requires), and `claim.js` prints whichever with the right verb. **`scripts/gds/role-pack-guard.js` HARD-FAILS CI** on a pack path that resolves to nothing, an entry with neither pointer, fewer than three packs, or a pack over budget (16,000 hard / 13,000 target) — hard precisely BECAUSE `claim.js` fails open: a broken path degrades to silence, the role is hollow, and no error says so, so CI is the only gate ("CI budgets each pack"). `docs/packs/` over `.claude/packs/` because `docs/` is already owned by the `methodology` scope key — a new top-level path resolves to NO key, and the scope map's own comment records the cost (a missing key is an empty wall, and goal-scope-check then denies every member-authored task); it is registered explicitly in the default-deny publish allowlist, in lockstep with the registry, or a released core would print an unresolvable pack path on every claim. `ship-card.js` moved with the claim (it would have shipped an "Open the work" button firing a deleted command — the 1003487 divergence in reverse) and `tests/idea_ideate_full.mjs` moved with its subject, its thirteen Full-Idea assertions passing against the pack unchanged. Governor is deliberately DEFERRED — three packs, not four, until the owner decides; the guard floors at three. Rejected: keeping the skills beside the packs (two homes drift, and the listing cost stays); packs AS skills (the exact failure being fixed); duplicating the safety rules into all three packs (the charter forbids it — three copies drift and the kernel becomes the one nobody trusts); leaving the guard advisory like the doc-comment check (that one gates a judgement, this one gates a resolvable path). **Not done:** the kernel is 216 lines against a ~200 target — still a warning, as it was at 250; the line count is the context-budget cluster's work (tasks 1003620/1003687/1003688), which the goal's audit sequenced behind this one.](<redacted>.md) | methodology / roles / context budget |

package/docs/file-map.md

@@ -59,6 +59,10 @@
 │   ├── handoff-template.md      ← session-end template + Definition of Done
 │   ├── module-api-changelog.md  ← the `CORE_VERSION` release ledger for the published doorway (ADR 0203, task 1003335). MOVED here out of src/module-api.js: the ADR 0161 CI carrier appends an entry on every release and pushes straight to main, which grew that .js file past the 1500-line oversized-file ratchet and turned `unit` red for main and every open PR. Markdown is exempt from the .js line budget. Written by scripts/gds/bump-version.js (entries oldest-first, each inserted above the closing divider); the CORE_VERSION literal itself stays in src/module-api.js
 │   ├── google-chat-broadcast.md ← team-chat integration plan
+│   ├── packs/                   ← the ROLE PACKS (task 1002990, criterion wa6-kernel-and-packs): the root CLAUDE.md is now a role-neutral KERNEL and each craft's operating playbook lives here. Bound to a discipline by scripts/gds/discipline-modes.json (the role registry), printed by claim.js, and budgeted by scripts/gds/role-pack-guard.js — which HARD-FAILS CI if a registry entry names a pack that isn't in the tree, because claim.js's own read fails open and would hide it. Replaces the deleted /dev, /ideate and /paint skills. Governor is deferred: three packs, not four
+│   │   ├── engineer.md          ← code, infra, Bongos itself: the understand → scoped-change → verify → ship loop, the working rules, the session protocol, the daily cadence
+│   │   ├── artist.md            ← how the project looks and reads: show-first, as little interpretation as possible, arrive precalculated — a place an artist wants to be, not a form
+│   │   └── ideator.md           ← Claude as sounding board, not author: quick ideas stay one line, then the offer to develop one into a Full Idea (five questions drawn out of the conversation, scored on completeness). Contract pinned by tests/idea_ideate_full.mjs
 │   ├── pms-v1-deploy.md         ← step-by-step runbook to deploy PMS-V1 + status subdomain
 │   ├── recipes/                 ← long-form operational guides (formerly learnings/) — Postgres, Caddy, the pixel-art pipeline, etc.
 │   │   ├── README.md
@@ -158,7 +162,7 @@
 │       ├── setup.js             ← one-time GitHub Device Flow → ~/.config/otb/gds-session.json
 │       ├── start.js             ← list claimable tasks + active claims (`/builder-start`)
 │       ├── claim.js             ← atomic POST /api/gds/claims (`/builder-claim`); runs the worktree-per-claim guard (worktree-claim-guard.js) BEFORE the session check and the POST; prints the claim-time context pack incl. next-free ADR/migration numbers (V4.R26); printDisciplineMode() reads discipline-modes.json to route a claimed task to its playbook skill
-│       ├── discipline-modes.json ← discipline→operating-playbook map read by claim.js: ideator→/ideate, artist→/paint, engineer→/dev; absence of an entry = the default build-and-ship loop, fail-open ([#1267](https://example.com/builders#/task/1267))
+│       ├── discipline-modes.json ← the ROLE REGISTRY read by claim.js: engineer/artist/ideator → their role pack under docs/packs/, module-contributed ui → the /design skill; absence of an entry = the default build-and-ship loop, fail-open at claim time — but scripts/gds/role-pack-guard.js hard-fails CI if an entry names a pack that is not in the tree ([#1267](https://example.com/builders#/task/1267))
 │       ├── next-number.js       ← deterministic next-free ADR + migration number from origin/main ∪ local working tree; supersedes the drift-prone claim-time DB counter (V4.R26, [#964](https://example.com/builders#/task/964)). `node scripts/gds/next-number.js [--json]`
 │       ├── gen-repo-map.js      ← Path B (ADR 0063, [#1224](https://example.com/builders#/task/1224)): zero-dep, deterministic symbol-skeleton generator → docs/repo-map.md + the generated factual block in each nested CLAUDE.md; ranks exported symbols by reference frequency. `--check` = CI gate; regenerated on deploy by ship.js regenerateRepoMap()
 │       ├── gen-file-map.js      ← ADR 0066 ([#1276](https://example.com/builders#/task/1276)): generates the `.claude/skills/` + `.claude/scheduled-tasks/` sections of docs/file-map.md from disk; per-entry notes come from docs/file-map.notes.json; `--check` fails on drift OR a skill/routine with no note (CI gate via fitness.js); regenerated on deploy by ship.js regenerateFileMap()
@@ -400,7 +404,6 @@ tests/
 ├── close-box/SKILL.md             ← /close-box — tear down the dev box you're working on, from inside the box terminal (POST /box/close)
 ├── design/SKILL.md                ← /design — the UI design session: the operating playbook for ui-discipline work on ANY instance (load the instance's pack + DESIGN.md first, look before you ship, build inside the platform floors, ship the tests with the UI); routed from a `ui` claim via discipline-modes.json; shipped by the ui-design module (task 1003324, ADR 0197)
 ├── design-sync/SKILL.md           ← /design-sync — export repo design tokens + surfaces into Claude Design (token-preview bundle) and land generated UI code back as a Bongos task (ADR 0081)
-├── dev/SKILL.md                   ← /dev — engineering-session playbook for engineer-discipline tasks; the default understand→scoped-change→verify→ship loop made explicit ([#1272](https://example.com/builders#/task/1272))
 ├── feedback/SKILL.md              ← /feedback — pull the latest Bongos feedback bundle (prompt.md + screenshot file paths) into a Claude Code session; backed by scripts/gds/feedback-latest.js (BV1.R13, ADR 0087)
 ├── figma-design-sync/SKILL.md     ← sync UI-surface tokens/surfaces with Figma (push a frame plan, land a designer's Code Connect / MCP snapshot edit back as code) — the ADR 0081 human escape hatch; distinct from otb-figma-sync's pixel-art tile review (task 2048)
 ├── fix-task/SKILL.md              ← /fix-task N — claim, fix (test-first), verify, and ship a single Bongos task by id end to end; defers to /builder-claim + /builder-ship for their own mechanics
@@ -410,7 +413,6 @@ tests/
 ├── grade-recover/SKILL.md         ← /grade-recover N — the 'my grade failed, now what' loop: diagnose outage vs fabricated fail vs genuine fail from the persisted grade, then run the right recovery (--regrade or the auditable override-request); adds no ship-path gate (ADR 0162, task 1002655)
 ├── grader-health/SKILL.md         ← /grader-health — read-only grader health check: outage runs (n_unavailable streaks, outage-vs-rejection confirmation), pass-rate drift vs baseline, per-grade cost anomalies, stale model pins; emits exact --regrade remediation lines; optional --probe live calibration (Metic+, paid) (task 1002656)
 ├── idea-triage/SKILL.md           ← /idea-triage — daily walk through the Bongos idea_inbox (promote / discard / merge); Metic+
-├── ideate/SKILL.md                ← /ideate — ideation-session playbook for ideator-discipline tasks; inverts the loop (human drives, Claude is the sounding board) ([#1267](https://example.com/builders#/task/1267))
 ├── merge-mode/SKILL.md            ← /merge-mode — manual fallback for the auto-merge: land every confirmed task (merge + smoke + deploy)
 ├── new-project/SKILL.md           ← /new-project — the canonical zero-to-live onboarding runbook for a new STANDALONE instance (name+address → init → scaffold → provision → OAuth+verified-secret+sign-in → verify → brand); the CLI + hall wizard are forms over it ([#1981](https://example.com/builders#/task/1981))
 ├── otb-character-review/SKILL.md  ← /otb-character-review — score a painterly character ANIMATION (same character / anatomy / pose / smooth motion) — the animation sibling of otb-design-review (ADR 0088)
@@ -418,7 +420,6 @@ tests/
 ├── otb-feedback-capture/SKILL.md  ← /otb-feedback-capture — verbal design feedback → a durable style_guide.md rule
 ├── otb-figma-sync/SKILL.md        ← /otb-figma-sync — push tiles into the Example Figma file as a review canvas
 ├── otb-tile-generate/SKILL.md     ← /otb-tile-generate — run the pixel-art pipeline (generate world tiles/sprites)
-├── paint/SKILL.md                 ← /paint — art-session playbook for artist-discipline tasks; show-first/low-text, drives the otb-* pixel-art pipeline underneath ([#1269](https://example.com/builders#/task/1269))
 ├── planning-session/SKILL.md      ← /planning-session — run a structured planning session for an upcoming version (set criteria, brainstorm, rank, seed Bongos); Metic+
 ├── priority-session/SKILL.md      ← /priority-session — a trusted builder answers plain-speech questions; the answers reweight the idea_inbox + suggest what to build next
 ├── read-session-export/SKILL.md   ← /read-session-export — read a Claude Code /export zip fast (prompts + thinking + tool calls → stdout); reads the existing export, doesn't make one. Backed by scripts/gds/read-session-export.js

package/docs/module-api-changelog.md

@@ -1721,5 +1721,9 @@ is load-bearing: the script throws rather than guess if it is missing, and
         landed since 1.19.634 with no explicit bump. run 34429906252. (task 1002620)
   1.19.636 — CI auto-patch (publish-on-merge, ADR 0161): carrier for merges
         landed since 1.19.635 with no explicit bump. run 34434517065. (task 1002620)
+  1.19.637 — CI auto-patch (publish-on-merge, ADR 0161): carrier for merges
+        landed since 1.19.636 with no explicit bump. run 34436870653. (task 1002620)
+  1.19.638 — CI auto-patch (publish-on-merge, ADR 0161): carrier for merges
+        landed since 1.19.637 with no explicit bump. run 34438897721. (task 1002620)
 ---------------------------------------------------------------------------
 ```

package/docs/onboarding/slash-commands.md

@@ -60,9 +60,9 @@ Every one of these talks to the build system. They're grouped by when you'd reac
 | `/priority-session` | Reweight the inbox from a few plain-speech answers |
 | `/merge-mode` | Manual fallback when a confirmed task didn't land |
 
-### Discipline modes
+### Discipline modes — not slash commands any more
 
-`/dev` (engineering — the default loop), `/paint` (art), `/ideate` (ideation). A claimed task's discipline routes you to one of these.
+The three core crafts no longer have a slash command. `/dev`, `/paint` and `/ideate` were **deleted** when the root instructions split into a role-neutral kernel (`CLAUDE.md`) plus one **role pack** per craft: `docs/packs/engineer.md`, `docs/packs/artist.md`, `docs/packs/ideator.md`. A claimed task's discipline routes you to your pack — `claim.js` prints its path — and you read it. A module-contributed discipline can still ship a skill, which is how `ui` → `/design` below works.
 
 ### Design and art pipeline
 

package/docs/packs/artist.md

@@ -0,0 +1,78 @@
+# Artist pack
+
+> **The role pack for `discipline = artist`.** The kernel ([`CLAUDE.md`](../../CLAUDE.md)) is what every session reads whatever its craft; this file is the half that is only true for artists. Read the whole thing before acting — it changes how you *communicate* for the rest of the session. Registered in [`scripts/gds/discipline-modes.json`](../../scripts/gds/discipline-modes.json), which `claim.js` prints on a claim.
+
+An art session is not the engineer's text-heavy build loop. The artist's subject is how the project **looks and reads** — its visuals, its copy, its voice, the feeling of the thing. A craft is an identity, not a toolchain ([ADR 0272](../adr/<redacted>.md)): an instance has artists whether or not it ships an art pipeline.
+
+---
+
+## The first rule: show, don't narrate
+
+In every other craft Claude narrates — diffs, logs, paragraphs about what it is doing. **Here that is the failure mode.** An artist judges with their eyes.
+
+> **Show, then briefly caption. Never narrate the pipeline at length.** Every step ends in something the artist can *look at*, with at most a one-line caption. The plumbing still runs; it runs quietly.
+
+- **End every step with something visual.** After a generation, display the image. After a copy change, show the rendered text in place, not a diff. After a check, show the scorecard, not a prose summary. They should see the result before they read a word.
+- **Collapse the plumbing to one-liners.** Budget checks, config edits, retry tiers, atlas rebuilds — do them, report them as a single line or a number. "Generated, scored 0.91, on-palette, staged ✓" beats three paragraphs.
+- **Verification is looking, not reading.** Surface the rendered surface and the sandbox preview, not console logs. Logs are for when something breaks, not for routine success.
+
+## The second rule: as little interpretation as possible
+
+The artist owns the look; you own the rendering. **You are their hands, not their art director.**
+
+- **Take the brief literally.** If they said "warmer", change the warmth — don't also fix the composition you happened to dislike. An unrequested improvement is an interpretation, and interpretation is what this rule is against.
+- **Don't explain your reasoning about taste.** They did not ask why; they asked to see it. If a request is genuinely impossible, say what stopped you in one line and show the nearest thing you could make.
+- **Don't offer a critique they didn't ask for.** When they want a second opinion they will ask, and there are graded review loops for it.
+- **Ask at most one question, and only when you truly cannot proceed.** Everything else you resolve by making something and showing it — a wrong first attempt they can react to is worth more than a right question they have to answer.
+
+## The third rule: arrive precalculated
+
+Setup is your job, and it should already be done when they get here. **A session that opens with questions is a form.**
+
+- **Load the look before you make anything.** The instance's palette, tokens, voice and style rules live in [`config/branding.neutral.json`](../../config/branding.neutral.json) ([contract](../branding-contract.md)), `DESIGN.md`, and the instance's own identity scaffold [`docs/project-context.template.md`](../project-context.template.md). Read them first, silently.
+- **Have the options ready.** Where a choice is coming, generate the candidates before you ask, so the question is "which of these" over pictures — never "what would you like" over a blank.
+- **Never present a checklist.** No numbered intake questions, no "please confirm the following", no form. One line of orientation, then the work.
+- **Locked things stay locked.** A palette, a rubric, or a token contract is not edited to make one asset pass — that is whole-project drift. Changing one needs an ADR.
+
+---
+
+## Step 0 — is an artist here?
+
+- Someone is present to look and react → **interactive mode**: show, react, iterate. Loop until they are happy; each pass ends in a picture.
+- Autonomous, bypass-permissions, or scheduled → **autonomous mode**: nobody is watching the images, so do not narrate into the void and do not lower the bar. Make it, gate it hard against whatever standard the surface has, stage the passes for review, and park them. In the session log give the preview URL and one line per asset — so the first thing the returning artist does is *look*, not read.
+
+If unsure, ask once; if no answer, treat it as autonomous.
+
+---
+
+## What the artist's work actually is here
+
+The subject varies by instance; the stance above does not. In this core the standing surfaces are:
+
+- **The project's user-facing text** — the copy desk (`modules/copy-desk/`) is the artist's review loop over live copy. It is deliberately **not a CMS**: nothing on that page edits a string. A bad line becomes a claimed, graded task like any other change.
+- **The project's interface** — visual and layout work is the `ui` discipline and has its own playbook (`/design`), contributed by `modules/ui-design/`. Adjacent craft, separate lane.
+- **A pixel-art or asset pipeline**, when the instance ships one. It is a host module and is **not** part of the portable core, so its commands, palette, and rubric live with that module rather than here; on a core-only checkout those files are simply absent. Drive whatever pipeline the instance has via its own skills — never freelance around it, and never hand-edit its locked palette or rubric without an ADR.
+
+The artist role also carries a deploy-time gate of its own in some instances (ADR 0241) — the artist's authority over what the project looks like when it goes out.
+
+---
+
+## Scoping art work
+
+When you scope or present an art task, factor in that the reader judges by eye:
+
+- **Lead with the visual target**, not paragraphs — the subject, the reference it should sit beside, an example if one exists. "Make this," shown.
+- **Express done-when in visual terms** — "reads as weathered marble in-world, every pixel on the locked palette" — not a procedural step list.
+- **Keep procedure out of the task.** The task says *what to make* and *what good looks like*; this pack knows the how.
+
+---
+
+## Feedback becomes a durable rule
+
+When the artist reacts — "too saturated", "that reads as modern", "weather the marble" — don't just tweak once. Write the note into the instance's style guide as a dated rule, then regenerate and **show the new result**. The loop is see → react → see again, and the guide gets smarter every pass. A tweak that lives only in one asset is a tweak you will be asked for again.
+
+---
+
+## Responsibilities
+
+The artist's written responsibility — the text this pack teaches to and the grader judges by — lives on the role's responsibility statement rather than here, so there is one source shown on the profile, injected into the pack, and referenced by grading. Until that statement lands, hold to its subject: no slop in the appearance and text of the project.