@ericrisco/rsc 0.1.50 → 0.1.52

package/skills/gamedev-pathing/SKILL.md

@@ -1,6 +1,6 @@
 ---
 name: gamedev-pathing
-description: "Use when implementing AI navigation or pathfinding in a game - moving enemies/NPCs/units toward goals, \"how do enemies navigate the level\", A*/Dijkstra/JPS/HPA* search, choosing grid vs waypoint-graph vs navmesh, baking a navmesh and tuning agent radius/height, off-mesh/nav links, dynamic obstacles and re-baking, steering (seek/flee/arrive/pursue/wander/path-following/flocking), local avoidance (RVO/ORCA, crowds), and flow fields for many agents to one goal. Per-engine: Godot 4.x (NavigationServer2D/3D, NavigationRegion, NavigationAgent, get_next_path_position), Unity (NavMeshAgent + NavMeshObstacle + AI Navigation baking), Unreal (RecastNavMesh + AIController + Behavior Trees/EQS). Triggers: \"my enemies get stuck\", \"units clump together\", \"navmesh not updating\", \"agentes se atascan\". NOT physics/rigidbody collision response or character-controller tuning (gamedev-physics), and NOT high-level enemy AI behaviour design like aggro/states/difficulty (game-design)."
+description: "Use when making NPCs, enemies, or units navigate a level — grid vs waypoint vs navmesh, A*/JPS, navmesh baking and agent radius, off-mesh links, steering, RVO crowd avoidance, and flow fields in Godot 4.x, Unity, Unreal. NOT collision response or character controllers (that is gamedev-physics), NOT aggro or difficulty design (that is game-design)."
 tags: [pathfinding, navmesh, navigation, steering, a-star]
 recommends: [godot, unity, unreal, gamedev-physics]
 profiles: [full]
@@ -102,7 +102,8 @@ Combine either as a **weighted sum** (simple) or **priority/arbitration** (avoid
 **Local avoidance (RVO / ORCA):** each agent picks a velocity that is collision-free assuming neighbours
 share the burden (*reciprocal* — hence no oscillating "dance"). This is avoidance, **not** collision
 *response*: it changes intended velocity *before* moving; the physics/collision solver is a separate,
-last-resort backstop (→ `gamedev-physics`). For dense crowds use the engine's crowd/avoidance system.
+last-resort backstop (→ [`gamedev-physics`](../gamedev-physics/SKILL.md)). For dense crowds use the
+engine's crowd/avoidance system.
 
 **Flow fields** — for **many agents → one (or few) goals** (RTS swarm, tower-defense creeps): run one
 **Dijkstra from the goal** over the grid to build a cost/integration field, derive a per-cell direction
@@ -182,30 +183,26 @@ AICon->MoveToActor(TargetActor, /*AcceptanceRadius*/ 50.f);   // or MoveToLocati
 4. Costs/holes: **`NavModifierVolume`** + `NavArea` classes. Off-mesh: **`NavLinkProxy`**. Crowd
    avoidance: enable the **Detour Crowd** manager (or `DetourCrowdAIController`) for RVO on many agents.
 
-> High-level BT *design* (states, aggro, difficulty) is `game-design`; this skill wires the BT's
+> High-level BT *design* (states, aggro, difficulty) is [`game-design`](../game-design/SKILL.md); this skill wires the BT's
 > movement/EQS tasks to navigation, not the decision tree's semantics.
 
-## Guardrails / gotchas
+## Anti-patterns
 
-- **Agent radius mismatch** is the top "stuck in doorways / clipping walls" cause — bake with the real radius.
-- Bake covers **static** geometry; runtime-spawned meshes need a rebake, a carving obstacle, or a nav link.
-- Path returns empty when start/end are **off the navmesh** or in disconnected islands — snap to nearest poly and check reachability first.
-- Avoidance (RVO) ≠ collision (physics). Agents can still overlap under pressure; that is expected, not a physics bug.
-- Off-mesh links are **directional and manual** — a jump-down link does not imply a jump-up link.
-- Smooth grid paths (funnel / string-pull) or agents walk visible zig-zag staircases.
-
-## Anti-patterns / rationalizations → STOP
-
-| Rationalization | Reality / do instead |
+| Anti-pattern | Do instead |
 | --- | --- |
-| "Just A* every frame, it's fine" | Melts CPU at scale; path on goal-change, steer between frames. |
-| "One navmesh bake works for every unit" | Radius/height differ; bake per agent size or use agent-type profiles. |
-| "Agents clip through each other — physics bug" | Missing **local avoidance**; enable RVO/crowd, that's the navigation layer. |
-| "Navmesh doesn't see my spawned wall" | Static bake; set Dynamic/rebake, or add a carving obstacle/nav link. |
-| "I'll grid-search my own A* in Unity/Unreal" | Reinvents the built-in navmesh; use `NavMeshAgent` / `AAIController::MoveTo`. |
-| "500 zombies each run A* to the player" | That's a **flow field**: one Dijkstra from the goal, agents sample cells. |
-| "Set `agent.destination` and read the path same frame" | Path is async (`pathPending`); gate on it before trusting `remainingDistance`. |
-| "Bigger heuristic = smarter" | Over-estimating `h` breaks A* optimality; keep it admissible (or weight it *knowingly*). |
+| Running a full A* every frame | Path on goal-change (or throttled) and steer between frames — per-frame search melts CPU at scale. |
+| One navmesh bake shared by every unit | Radius/height differ; bake per agent size or use agent-type profiles. |
+| Baking with a placeholder agent radius | Bake with the real radius — mismatch is the top "stuck in doorways / clipping walls" cause. |
+| Treating agents that clip through each other as a physics bug | Missing **local avoidance**; enable RVO/crowd — that is the navigation layer. |
+| Expecting RVO to prevent every overlap | Avoidance ≠ collision. Agents can still overlap under pressure; that is expected, not a physics bug. |
+| Expecting a **static** bake to see runtime-spawned geometry | Set Dynamic runtime generation / rebake, or add a carving obstacle or nav link. |
+| Hand-rolling your own grid A* in Unity/Unreal | Reinvents the built-in navmesh; use `NavMeshAgent` / `AAIController::MoveTo`. |
+| 500 zombies each running A* to the player | That's a **flow field**: one Dijkstra from the goal, agents sample cells. |
+| Setting `agent.destination` and reading the path the same frame | Path is async (`pathPending`); gate on it before trusting `remainingDistance`. |
+| Inflating `h` because a bigger heuristic seems smarter | Over-estimating `h` breaks A* optimality; keep it admissible (or weight it *knowingly*). |
+| Trusting a path query that came back empty | Start/end are **off the navmesh** or in disconnected islands — snap to the nearest poly and check reachability first. |
+| Assuming an off-mesh link works both ways | Links are **directional and manual** — a jump-down link does not imply a jump-up link. |
+| Feeding raw grid paths to the mover | Smooth them (funnel / string-pull) or agents walk visible zig-zag staircases. |
 
 ## Related skills
 
@@ -213,7 +210,7 @@ AICon->MoveToActor(TargetActor, /*AcceptanceRadius*/ 50.f);   // or MoveToLocati
 - [`unity`](../unity/SKILL.md) — Unity/C# project setup; here for the NavMesh + AI Navigation package details.
 - [`unreal`](../unreal/SKILL.md) — UE5/Blueprint/C++; here for RecastNavMesh + AIController/BT/EQS wiring.
 - [`gamedev-physics`](../gamedev-physics/SKILL.md) — collision *response*, rigidbodies, character controllers; steering decides intended velocity, physics resolves the contact.
-- `game-design` — high-level enemy behaviour, aggro, encounter/difficulty design (what the AI *decides*, not how it *moves*).
+- [`game-design`](../game-design/SKILL.md) — high-level enemy behaviour, aggro, encounter/difficulty design (what the AI *decides*, not how it *moves*).
 
 ## Checklist
 
@@ -234,9 +231,3 @@ the project's navigation decisions in `02-DOCS/wiki/stack/gamedev-pathing.md` (i
 `02-DOCS/wiki/index.md`): engine + version, chosen representation, bake settings (agent sizes, cell
 size), avoidance/crowd choice, custom links/areas. Read it first on every use; bump its `Updated` date
 when a convention changes. No `02-DOCS/` layer? Skip silently — conventions are *recorded, not gated*.
-
-## See Also — local references (read when)
-
-- `references/search-algorithms.md` — representations, A*/Dijkstra/JPS/HPA* pseudocode, heuristics & admissibility, tie-breaking, any-angle, funnel smoothing.
-- `references/navmesh-workflow.md` — bake-parameter table, areas/costs, off-mesh links, obstacles vs carving vs partial rebake, per-engine baking, debugging.
-- `references/steering-and-avoidance.md` — steering primitives, path following, boids weights, RVO/ORCA intuition, crowds, flow-field construction.

package/skills/gamedev-shipping/SKILL.md

@@ -1,6 +1,6 @@
 ---
 name: gamedev-shipping
-description: "Use when exporting, building, packaging, or publishing a game — per engine (Godot export templates, Unity Build Profiles, Unreal Shipping vs Development) and per platform: desktop (Windows/macOS/Linux; macOS codesign + notarytool notarization), web (the COOP same-origin + COEP require-corp headers threaded SharedArrayBuffer WebAssembly builds need, and that Godot's C#/.NET edition can't export to web), mobile (Android JDK 17 build template, AAB vs APK, release keystore; iOS provisioning). Also store prep (icons/splash, metadata, age ratings, AAB for Play, Mac App Store, Steamworks/SteamPipe, itch.io butler), headless-export CI/CD, build/version numbers, and ship-time size/startup optimization. Triggers: 'export my game', 'build for Steam', 'ship to web/mobile', 'publish to the Play Store', 'notarize my Mac build', 'itch.io upload', 'SharedArrayBuffer is not defined'. NOT for writing gameplay/runtime code (that's the engine skill — godot/unity/unreal) or hosting a normal web app/API/container (→ deployment)."
+description: "Use when exporting, packaging, signing, or publishing a game — export presets per engine, macOS notarization, cross-origin isolation for threaded web builds, Android AAB/keystore, iOS provisioning, store submission (Steam, Play, App Store, itch.io), headless build CI. NOT gameplay code inside the artifact (that is `godot`, `unity`, `unreal`), NOT hosting a web app or container (that is `deployment`)."
 tags: [export, build, release, packaging, store, steam, itch, notarization, webassembly, aab]
 recommends: [godot, unity, unreal, deployment]
 profiles: [full]
@@ -29,21 +29,14 @@ Godot templates·Unity Build Profiles·Unreal Shipping    desktop·web·mobile
 | Android | JDK **17**, target a recent API level | AAB required for Google Play (new apps + updates). APK only for sideload/itch/direct. |
 | iOS/macOS | Xcode 16+, `notarytool` | `altool` notarization is dead — use `xcrun notarytool`. |
 
-## When to use / When NOT to use
+## When NOT to use — and who owns it instead
 
-**Use when:**
-- Configuring export templates / build presets / packaging settings for any target.
-- Signing & notarizing a desktop build, or debugging Gatekeeper/SmartScreen rejections.
-- Fixing a web build that boots to `SharedArrayBuffer is not defined` or a blank canvas.
-- Producing an AAB/APK or an iOS archive; wiring keystores/provisioning.
-- Store submission prep (icons, metadata, ratings) or uploading to Steam/itch/Play/App Store.
-- Automating headless exports in CI, or setting version/build numbers.
-- Shrinking build size / startup time before release.
-
-**NOT for (route elsewhere, say so, stop):**
-- Writing gameplay, shaders, netcode, or any runtime logic → the **engine** skill (`godot`/`unity`/`unreal`) or `gamedev-shaders`/`gamedev-multiplayer`. This skill packages code; it does not write it.
-- Hosting a normal web app, API, or container (Dockerfile, Coolify, Vercel, VPS) → **`deployment`**. A WebAssembly game is *static files*; only its **headers** are special (below).
-- Designing game feel, economy, or narrative → `game-design` / `game-storytelling`.
+| Ask | Owner | Why the line is there |
+| --- | --- | --- |
+| Gameplay, shaders, netcode, physics — any runtime logic | `../godot/SKILL.md`, `../unity/SKILL.md`, `../unreal/SKILL.md`, `../gamedev-shaders/SKILL.md`, `../gamedev-multiplayer/SKILL.md`, `../gamedev-physics/SKILL.md` | This skill packages code, it does not write it. The boundary is the export button. |
+| Hosting a normal web app, API or container (Dockerfile, Coolify, Vercel, VPS) | `../deployment/SKILL.md` | A WebAssembly game is *static files*; only its **headers** are special (below). Borrow its Actions + header-setting patterns. |
+| Game feel, economy, narrative | `../game-design/SKILL.md`, `../game-storytelling/SKILL.md` | Design intent, not packaging. |
+| Signing-key and CI secret handling in general | `../secure-coding/SKILL.md` | The practice this skill assumes for keystores/certs. |
 
 ## Decision rules (settle these before touching a build)
 
@@ -96,7 +89,8 @@ RunUAT BuildCookRun -project="Game.uproject" -platform=Win64 \
 Android adds `-distribution` for an AAB; iOS requires a Mac + provisioning. Run the platform
 **Turnkey**/SDK setup once (Android Studio + NDK, or Xcode) or the cook fails.
 
-→ full per-engine preset/flag tables: `references/mobile-and-desktop.md` and `references/stores-and-ci.md`.
+→ full per-engine preset/flag tables: [references/mobile-and-desktop.md](references/mobile-and-desktop.md)
+and [references/stores-and-ci.md](references/stores-and-ci.md).
 
 ## Per-platform
 
@@ -141,7 +135,8 @@ files are fine. Where to set them:
 `COEP: require-corp` also means every cross-origin subresource (fonts, analytics, embeds) must
 send CORP/CORS or it's blocked. **Godot C#/.NET has no web export at all** — use GDScript.
 
-→ hosts, exact configs, the service-worker shim, self-test: `references/web-export-headers.md`.
+→ per-host configs, the service-worker shim, COEP subresource fallout, a self-test snippet and the
+Godot vs Unity web specifics: [references/web-export-headers.md](references/web-export-headers.md).
 
 ### Mobile
 
@@ -162,7 +157,8 @@ Keep the keystore + passwords out of git (CI secret). `versionCode` must increas
 then upload to App Store Connect / TestFlight via Xcode Organizer or **Transporter**. `CFBundleVersion`
 must increase per upload.
 
-→ full keystore, Play App Signing, provisioning, TestFlight walkthrough: `references/mobile-and-desktop.md`.
+→ full keystore + Play App Signing, provisioning/TestFlight, Windows Authenticode and Linux
+AppImage/Flatpak walkthroughs: [references/mobile-and-desktop.md](references/mobile-and-desktop.md).
 
 ## Store prep
 
@@ -174,7 +170,8 @@ must increase per upload.
 - **itch.io**: `butler push build/ user/game:channel --userversion 1.2.3` — channel name (e.g. `windows`, `html5`) picks the platform.
 - **Mac App Store**: Apple Distribution cert + App Sandbox entitlements (not Developer ID); notarization is handled by App Store submission.
 
-→ Steamworks/SteamPipe VDF, butler channels, Play/App Store checklists: `references/stores-and-ci.md`.
+→ Steamworks/SteamPipe VDF, butler channels, Play/App Store submission checklists:
+[references/stores-and-ci.md](references/stores-and-ci.md).
 
 ## CI/CD for builds
 
@@ -186,7 +183,7 @@ Headless export in GitHub Actions; version from the run/commit so every artifact
 - **Signing in CI**: import the keystore/cert from an encrypted secret at runtime, never commit it. macOS notarization runs on a `macos-latest` runner with `notarytool` + an App Store Connect API key.
 - **Versioning**: semver tag for releases; a monotonic build number from `github.run_number` or the git commit count feeds `versionCode`/`CFBundleVersion`/Steam build description.
 
-→ ready workflows per engine: `references/stores-and-ci.md`.
+→ ready per-engine workflows with signing: [references/stores-and-ci.md](references/stores-and-ci.md).
 
 ## Ship-time optimization
 
@@ -195,26 +192,19 @@ Headless export in GitHub Actions; version from the run/commit so every artifact
 - **Startup**: small boot scene, stream/preload heavy assets, compress the `.pck`/data, use Ogg Vorbis over WAV. Web: enable gzip/brotli on `.wasm`/`.pck`.
 - **Budgets**: set a per-platform size budget (web especially — every MB is download latency) and check the final artifact against it before submitting.
 
-## Anti-patterns — rationalization → STOP
+## Anti-patterns
 
-| Rationalization | STOP — do this instead |
+| Anti-pattern | Do instead |
 | --- | --- |
-| "Web build works locally, host later" | Set COOP/COEP now; `file://`/simple servers hide the `SharedArrayBuffer` failure. |
-| "I'll add C# web export for Godot" | It doesn't exist in 4.x. GDScript for web, or no web. |
-| "Notarization is optional" | On macOS, no notarize+staple = Gatekeeper blocks the app for every user. |
-| "Ship an APK to the Play Store" | Play requires **AAB** for new apps and updates. |
-| "Commit the keystore so CI can sign" | Keystore/cert live in encrypted CI secrets, never in git. |
-| "Ship the Development/Debug config" | Ship Godot release / Unity IL2CPP release / Unreal **Shipping** — debug leaks console + is slow. |
-| "Templates are close enough (4.5 vs 4.5.1)" | Godot templates must match the editor build exactly or export breaks. |
-| "Uncompressed textures are fine" | Per-platform compression (ASTC/BCn/ETC2) — raw textures blow memory + size. |
-| "Reuse an old versionCode/CFBundleVersion" | Both must strictly increase or the store rejects the upload. |
-
-## Related skills
-
-- `godot`, `unity`, `unreal` — write the game; this skill exports/ships what they build. Draw the line at the export button.
-- `deployment` — Docker/CI/hosting for **web apps & APIs**; borrow its GitHub Actions + header-setting patterns, but a WebAssembly game is static files, not a container.
-- `gamedev-shaders`, `gamedev-multiplayer`, `gamedev-physics` — runtime subsystems packaged into the ship artifact, not configured here.
-- `secure-coding` — signing-key and secret handling this skill assumes for keystores/certs in CI.
+| Web build works locally, host it later | Set COOP/COEP now; `file://` and simple servers hide the `SharedArrayBuffer` failure. |
+| Planning a Godot C# web export | It doesn't exist in 4.x. GDScript for web, or no web. |
+| Treating notarization as optional | On macOS, no notarize + staple = Gatekeeper blocks the app for every user. |
+| Shipping an APK to the Play Store | Play requires **AAB** for new apps and updates. |
+| Committing the keystore so CI can sign | Keystore/cert live in encrypted CI secrets, never in git. |
+| Shipping the Development/Debug config | Ship Godot release / Unity IL2CPP release / Unreal **Shipping** — debug leaks console + is slow. |
+| Export templates "close enough" (4.5 vs 4.5.1) | Godot templates must match the editor build exactly or export breaks. |
+| Uncompressed textures | Per-platform compression (ASTC/BCn/ETC2) — raw textures blow memory + size. |
+| Reusing an old versionCode/CFBundleVersion | Both must strictly increase or the store rejects the upload. |
 
 ## Checklist
 
@@ -228,9 +218,3 @@ Headless export in GitHub Actions; version from the run/commit so every artifact
 - [ ] Store assets: icons, splash, screenshots, metadata, age rating done.
 - [ ] Version/build number set from CI and traceable.
 - [ ] Textures compressed per platform; build size within budget.
-
-## References
-
-- `references/web-export-headers.md` — cross-origin isolation deep dive: COOP/COEP per host (Netlify/Vercel/nginx/itch/GitHub Pages), the `coi-serviceworker` shim, COEP subresource fallout, self-test snippet, Godot vs Unity web specifics.
-- `references/mobile-and-desktop.md` — Android keystore + Gradle template + Play App Signing; iOS provisioning/TestFlight; full macOS codesign→notarytool→staple; Windows Authenticode; Linux AppImage/Flatpak.
-- `references/stores-and-ci.md` — Steamworks/SteamPipe VDF upload, itch.io butler channels, Play/App Store submission checklists, and per-engine GitHub Actions headless-build workflows with signing.

package/skills/godot/SKILL.md

@@ -1,6 +1,6 @@
 ---
 name: godot
-description: "Use when writing, reviewing, or debugging Godot 4.x games in GDScript (typed) or C# — nodes/scenes/custom Resources, autoload singletons and an EventBus, signals via Callable, @onready/@export/@tool, static typing, CharacterBody2D/3D + velocity + move_and_slide(), _process vs _physics_process, .tscn/.tres text formats, and headless GUT/gdUnit4 tests. Enforces Godot 4 APIs and rejects Godot 3 leftovers: yield→await, onready→@onready, .instance()→.instantiate(), KinematicBody→CharacterBody, Spatial→Node3D, string connect(\"sig\",obj,\"m\")→Callable. Triggers: 'write a Godot player controller', 'my move_and_slide does nothing', 'port this Godot 3 script to 4', 'connect a signal in Godot', 'GDScript won't type-check', 'CharacterBody2D won't move', 'add an autoload', 'escribe un script de Godot', 'por qué no se mueve mi personaje'. NOT Unity MonoBehaviour or Unreal (route to unity/unreal); NOT pure C++/CMake tooling unrelated to GDExtension (that is cpp); NOT shader code in .gdshader (that is gamedev-shaders)."
+description: "Use when writing, reviewing, or debugging Godot 4.x games in GDScript or C# — scenes, nodes, custom Resources, autoload/EventBus signals, typed GDScript, CharacterBody2D/3D movement, GUT/gdUnit4 — or porting Godot 3 APIs to 4. NOT Unity/Unreal (that is `unity`/`unreal`), NOT `.gdshader` code (`gamedev-shaders`), NOT GDExtension C++ tooling (`cpp`)."
 tags: [godot, gdscript, game-dev, csharp, godot4]
 recommends: [gamedev-shaders, gamedev-multiplayer, gamedev-physics, gamedev-pathing, gamedev-shipping, game-design, cpp]
 profiles: [full]
@@ -9,9 +9,9 @@ origin: risco
 
 # Godot 4.x (GDScript + C#)
 
-Write, review, and debug Godot **4.x** games — 2D and 3D — the way the engine is designed:
-scenes as reusable units, composition over deep node trees, signals for decoupling, static
-typing for speed and safety. GDScript (typed) is primary; C# parity snippets sit beside it.
+Build 2D and 3D the way the engine is designed: scenes as reusable units, composition over deep node
+trees, signals for decoupling, static typing for speed and safety. GDScript (typed) is primary; C#
+parity snippets sit beside it.
 
 ## Version contract — read first
 
@@ -111,8 +111,11 @@ Keep gameplay logic in nodes; let the bus carry the *notification*, not the beha
   children in `_init`.
 - **Prefer unique names**: mark a node **Unique Name in Owner** (`%`) and access `%HealthBar`
   instead of the fragile, refactor-breaking `get_node("../../UI/HealthBar")`. `$Foo` is fine for a
-  direct child.
+  direct child — `get_node`/`$` on a missing path returns `null` and errors.
 - Cache node lookups in `@onready` vars; don't call `get_node` every frame.
+- **Freeing**: call `queue_free()` (safe, end of frame), not `free()` mid-signal. Guard reused refs
+  with `is_instance_valid(node)`.
+- Never busy-wait; `await get_tree().create_timer(1.0).timeout` or `await` a signal.
 
 ```gdscript
 extends CharacterBody2D
@@ -286,34 +289,12 @@ like any other library. That is native-tooling territory → pair with `cpp`.
 - Validate a project before shipping: open in the editor once (catches broken `.tscn`/`.tres`),
   then export-check per platform. Details → `references/export-and-testing.md`.
 
-## Guardrails / gotchas
-
-- Freeing: call `queue_free()` (safe, end of frame), not `free()` mid-signal. Guard reused refs with
-  `is_instance_valid(node)`.
-- Never busy-wait; `await get_tree().create_timer(1.0).timeout` or a signal.
-- `get_node`/`$` on a missing path returns `null` and errors — prefer `%UniqueName` and typed `@onready`.
-- Don't multiply by `delta` twice (see `move_and_slide`). Don't put physics in `_process`.
-- One `TileMapLayer` per layer; `TileMap` is deprecated.
-- Editor-touching code must guard on `Engine.is_editor_hint()` under `@tool`.
-
-## Related skills
-
-- [`game-design`](../game-design/SKILL.md) — mechanics, loops, and feel *before* you script them.
-- [`gamedev-shaders`](../gamedev-shaders/SKILL.md) — `.gdshader` / visual shaders (out of scope here).
-- [`gamedev-physics`](../gamedev-physics/SKILL.md) — joints, RigidBody tuning, deep collision layers.
-- [`gamedev-pathing`](../gamedev-pathing/SKILL.md) — NavigationAgent, A*, steering beyond the basics.
-- [`gamedev-multiplayer`](../gamedev-multiplayer/SKILL.md) — `MultiplayerSynchronizer`/RPC/netcode.
-- [`gamedev-shipping`](../gamedev-shipping/SKILL.md) — store builds, signing, platform export at scale.
-- [`cpp`](../cpp/SKILL.md) — the GDExtension / `godot-cpp` native side and its CMake/build tooling.
-
-## Checklist
-
-- [ ] Zero Godot 3 APIs — checked every line against the ban-list (`yield`/`onready`/`instance()`/`KinematicBody`/`Spatial`/string `connect`/`File`).
-- [ ] `move_and_slide()` called with **no args**; `velocity` set as a property; not multiplied by `delta` twice.
-- [ ] Overridden lifecycle methods call `super()` where a base defines them.
-- [ ] Signals connected via Callable (`x.sig.connect(_on_x)`), named past-tense, disconnected/one-shot where needed.
-- [ ] Everything typed (`:` / `:=`, typed params/returns, typed Arrays); no stray `Variant`.
-- [ ] Physics in `_physics_process`, visuals in `_process`; rate-based values scaled by `delta`.
-- [ ] Node access via `@onready` + `%UniqueName`, not deep `get_node("../..")` chains.
-- [ ] `.tscn`/`.tres` not hand-corrupted; project opens in-editor before export.
-- [ ] C# snippets use `partial`, `[Export]`/`[Signal]`, `base._Ready()`, PascalCase (if C# requested).
+## Hand off to
+
+Mechanics, loops, and feel *before* you script them → [`game-design`](../game-design/SKILL.md).
+`.gdshader` / visual shaders → [`gamedev-shaders`](../gamedev-shaders/SKILL.md). Joints, RigidBody
+tuning, deep collision layers → [`gamedev-physics`](../gamedev-physics/SKILL.md). NavigationAgent,
+A*, steering → [`gamedev-pathing`](../gamedev-pathing/SKILL.md). `MultiplayerSynchronizer`/RPC/netcode
+→ [`gamedev-multiplayer`](../gamedev-multiplayer/SKILL.md). Store builds, signing, platform export at
+scale → [`gamedev-shipping`](../gamedev-shipping/SKILL.md). The GDExtension / `godot-cpp` native side
+and its CMake/build tooling → [`cpp`](../cpp/SKILL.md).

package/skills/grants/SKILL.md

@@ -1,6 +1,6 @@
 ---
 name: grants
-description: "Use when the goal is non-dilutive public or foundation funding — money you neither repay nor give equity for — and you must find a fit call, decide go/no-go, register an eligible entity, or write the application a reviewer scores against a published rubric (needs statement, logic model, SMART objectives, impact narrative, budget justification, LOI). Symptoms: a call with an Excellence/Impact/Implementation grid, a 3/5-per-criterion threshold, an indirect-cost line, a SAM.gov/UEI or BDNS lookup, an LOI that filters you before the full proposal. Triggers: 'find grants we qualify for', 'write the impact section so it scores above threshold', 'should we even bother applying to this call', 'build the budget justification', 'write a Letter of Inquiry', 'buscar la convocatòria a la BDNS', 'redactar la solicitud de subvención'. NOT equity rounds or investor pitches (that is fundraising), NOT a paying-customer SOW (that is proposals)."
+description: "Use when pursuing non-dilutive public or foundation funding — money you neither repay nor give equity for: finding a fit call, go/no-go, registration (UEI/PIC/BDNS), or writing the rubric-scored application, logic model, budget justification, or LOI. NOT an equity round (that is `fundraising`), NOT a paying-customer SOW (that is `proposals`)."
 tags: [grants, subsidies, fundraising-nondilutive, public-funding, grant-writing]
 recommends: [fundraising, proposals, financial-model, pitch-deck, bookkeeping, contracts]
 origin: risco
@@ -12,19 +12,7 @@ origin: risco
 
 You are a **reviewer-minded application writer**. Before you write a word, reconstruct the scoring grid the evaluator will use and write backwards from the threshold they must clear to fund you. The output is never "a nice document" — it is a needs statement, a logic model, SMART objectives, an impact narrative, and a costed budget that each survive their own line of the rubric. Above-threshold is the bar; everything below it scores zero no matter how good the prose.
 
-## When to use / When NOT to use
-
-Use when:
-
-- Building a fit shortlist of grants, public subsidies, or foundation calls an org qualifies for.
-- Deciding **go/no-go** on a specific call — eligibility, fit, effort vs. expected value, deadline vs. registration lead time.
-- Writing the application against a published evaluation rubric: needs, logic model, objectives, workplan, impact, budget justification.
-- Writing a **Letter of Inquiry / concept note** to a foundation.
-- Building the **logic model / theory of change / SMART objectives**, or the impact section.
-- Writing and justifying a **budget** so it survives allowability and indirect-cost rules.
-- Registering an eligible entity: US UEI via SAM.gov, EU PIC, finding the Spanish BDNS convocatoria.
-
-Do NOT use when — route instead:
+## Route elsewhere
 
 | The ask | Owner |
 |---|---|
@@ -36,8 +24,6 @@ Do NOT use when — route instead:
 | Post-award ledgers, fund accounting, spend tracking | `bookkeeping` |
 | Pricing, packaging, what to charge customers | `pricing` |
 
-The one-sentence boundary: `grants` owns the pursuit and writing of money scored by a **reviewer against a published rubric**; the moment it is **equity or a round** it is `fundraising`, and the moment the document persuades a **paying customer** it is `proposals`.
-
 ## The non-dilutive landscape
 
 Three lanes, and the rules live in different documents in each. Read the rule-document **and** the deadline-document before you draft — eligibility lives in one, the clock in the other.
@@ -162,10 +148,6 @@ Hard caps are real and enforced: Horizon Europe Part B is **40 pages** for stand
 | Generic LOI with no fit argument | LOIs filter on alignment first | Lead with funder-fit, then credibility, then impact |
 | Indirect line guessed or padded | Breaches the 15% de minimis with no negotiated rate | Use ≤15% MTDC, or declare your negotiated rate |
 
-## Jurisdiction quick-reference
-
-EU (Funding & Tenders portal, PIC, Horizon evaluation criteria, page limits), US (Grants.gov, SAM.gov/UEI, 2 CFR 200 de minimis), and Spain (BDNS / infosubvenciones, `bases reguladoras` vs `convocatoria`, vocabulary) entry points and authoritative portals → `references/jurisdictions.md`.
-
 ## Optional self-check
 
 `scripts/verify.sh path/to/application.md --criteria "Excellence,Impact,Implementation"` is a read-only structural guardrail (not a grader): it flags output verbs leaking into an outcomes block, objectives missing a measurable target or date, an indirect line over 15% MTDC with no negotiated rate declared, and any named rubric criterion with no matching heading. It exits 0 on a clean or empty file. The real rigor is the capability eval, not the linter.

package/skills/harness/SKILL.md

@@ -1,6 +1,6 @@
 ---
 name: harness
-description: "Use to control, govern or maintain a workspace's harness — software or non-code (a company, an ops desk, a knowledge vault). The harness is the control plane: the `01-TOOLS/` tooling layer, the `02-DOCS/` chaos→knowledge wiki, and the root Knowledge map. It audits, migrates legacy `XX-*` folders, scaffolds provider tooling, sweeps the inbox, and generates root CLAUDE.md/AGENTS.md. Brownfield-first. NOT the bootstrap front door (that is `init`, which hands off here)."
+description: "Use when governing a workspace's control plane, code or not — the `01-TOOLS/` tooling layer, the `02-DOCS/` chaos→knowledge wiki, the root Knowledge map. Audits it, migrates legacy `XX-*` folders, scaffolds provider tooling, sweeps the inbox, writes root CLAUDE.md/AGENTS.md. NOT the bootstrap front door (that is `init`, which hands off here)."
 tags: [harness, company, ops, docs, wiki, connect, tools, knowledge]
 recommends: [init]
 profiles: [minimal, core, full]
@@ -18,8 +18,9 @@ The **harness** is the control plane of a workspace. A workspace need not be cod
 
   Two on-ramps feed it. **Ingest**: the user drops any file in any format into `inbox/`, and the
   Auto-Ingest Sweep extracts, classifies, cross-links and compiles it — then goes for a walk,
-  discovering un-ingested documents anywhere in the workspace, bounded by `.rscignore` and
-  de-duplicated through the `wiki/.ingested.json` ledger. **Worklog**: every meaningful session of
+  discovering un-ingested documents anywhere in the workspace, bounded by `.rscignore` (baseline:
+  `references/ingest-ignore-defaults.md`) and de-duplicated through the `wiki/.ingested.json`
+  ledger. **Worklog**: every meaningful session of
   work is itself a raw source in `raw/worklog/`, captured on `PreCompact`/`SessionEnd`, at a commit
   milestone, or by the daily curation pass.
 
@@ -35,7 +36,8 @@ The **harness** is the control plane of a workspace. A workspace need not be cod
 
   It compounds on its own: every Ingest, Sweep and Query triggers a Maintenance Pass (lint, score
   recomputation, gap detection, Related sweep), a Micro-Improve runs every N interactions, and Deep
-  Improve runs on request or on the daily schedule. Protocol → `references/wiki-protocol.md`;
+  Improve runs on request or on the daily schedule (`references/daily-curation-automation.md` — on
+  Claude Code, wire it via the `schedule` skill). Protocol → `references/wiki-protocol.md`;
   formats → `references/ingest-formats.md`; capture → `references/wiki-worklog-template.md`;
   vault → `references/obsidian-scaffolding.md`.
 - **The Knowledge map** — the `## Knowledge map` section of the root `CLAUDE.md` that indexes the wiki (including the `harness/` topic) and is read by every other skill before it works in its area.
@@ -78,45 +80,7 @@ for harness-level choices, but defer concrete deploy mechanics to `deployment`,
 
 The skill proposes, the user confirms, the skill executes. Every destructive operation (deleting a legacy folder, merging into an existing `CLAUDE.md`) requires explicit consent quoted back from the user, not inferred.
 
-## When to use
-
-- A workspace has grown organically and needs the canonical `01-TOOLS` / `02-DOCS` structure.
-- Old numbered folders (`00-TOOLS`, `03-NOTES`, `04-LEGACY`…) exist and need to be consolidated.
-- A new project is being kicked off and needs the harness from day one.
-- Provider integrations exist in code but there's no operational tooling for them.
-
-**Do NOT use when:**
-- The user only wants to add a single tool — they can `cp -r 01-TOOLS/_TEMPLATE 01-TOOLS/<X>` manually.
-- The user wants to refactor runtime code — this skill is operational tooling only, never runtime.
-
-## Architecture
-
-```
-harness/
-├── SKILL.md                          ← this file (the protocol)
-├── references/
-│   ├── providers.yaml                ← detector catalog with full file bodies per provider
-│   ├── claude-md-template.md         ← root CLAUDE.md template
-│   ├── agents-md-template.md         ← root AGENTS.md template
-│   ├── tools-readme-template.md      ← 01-TOOLS/README.md catalog template
-│   ├── audit-report-template.md      ← exact format of the audit report shown to user
-│   ├── wiki-protocol.md              ← embedded wiki protocol + Inbox/Worklog Sweep + Continuous Improvement
-│   ├── ingest-formats.md             ← multiformat Fetch (PDF, image, CSV, JSON, html…)
-│   ├── ingest-ignore-defaults.md     ← what the sweep skips by default (the .rscignore baseline)
-│   ├── inbox-readme-template.md      ← the inbox/README.md drop-zone contract
-│   ├── wiki-raw-template.md          ← format for raw/<topic>/*.md
-│   ├── wiki-worklog-template.md      ← format for raw/worklog/*.md (work-driven capture)
-│   ├── wiki-article-template.md     ← format for wiki/<topic>/*.md (OKF frontmatter + markdown links)
-│   ├── wiki-index-template.md       ← format for wiki/index.md (machine catalog; .base = human nav)
-│   ├── obsidian-scaffolding.md      ← .base views, .gitignore, attachments, .obsidian config
-│   ├── daily-curation-automation.md ← portable scheduled compounding pass (the third trigger)
-│   ├── wiki-archive-template.md     ← format for archived query answers
-│   └── wiki-gaps-template.md        ← format for wiki/gaps.md (Knowledge Gaps log)
-├── assets/
-│   └── _TEMPLATE/                    ← per-tool boilerplate (cp -r seed)
-└── examples/
-    └── audit-example.md              ← walked-through audit on a synthetic project
-```
+**Out of scope:** adding a single tool — the user does `cp -r 01-TOOLS/_TEMPLATE 01-TOOLS/<X>` manually, no need for the full protocol; and refactoring runtime code — this skill is operational tooling only, never runtime.
 
 ## Protocol — five phases
 
@@ -148,7 +112,7 @@ Walk the workspace root and gather:
 
 Render **two artifacts**:
 
-1. **A compact text summary in the conversation** — 1–3 sentences per section, the full destructive-ops list, the consent prompt. This keeps the terminal flow fast.
+1. **A compact text summary in the conversation** — 1–3 sentences per section, the full destructive-ops list, the consent prompt, in the format of `references/audit-report-template.md`. This keeps the terminal flow fast. A full walked-through audit on a synthetic project: `examples/audit-example.md`.
 2. **A full HTML report at `<workspace_root>/02-DOCS/audits/audit-YYYY-MM-DD-HHMM.html`** using `references/audit-report-template.html`. Self-contained (inline CSS, no CDN). Includes color-coded action tables, collapsible legacy-folder sections, highlighted destructive ops, and the consent prompt. **Gitignored** (per-run artifact).
 
 If `02-DOCS/audits/` does not exist, create it (with `.gitkeep`) before writing — even on first run, before Phase 4 builds the rest of `02-DOCS/`. Same for `02-DOCS/` itself: the audits subdirectory is the only piece allowed to materialize during Phase 2; the rest waits until APPLY. Never write the audit HTML at the workspace root.
@@ -184,7 +148,7 @@ Execute in this exact order. Each step writes to disk; abort and report on first
 1. **Root files.**
    - If `CLAUDE.md` does not exist: render `references/claude-md-template.md` with the scan data and write it.
    - If `CLAUDE.md` exists: read it, compute a section-level diff against the template, and apply ONLY additive merges. Never delete user content. Never overwrite a section the user has customized. Append missing sections at the end with an `<!-- added by harness YYYY-MM-DD -->` marker.
-   - Same logic for `AGENTS.md`.
+   - Same logic for `AGENTS.md`, rendered from `references/agents-md-template.md`.
 2. **Create `01-TOOLS/` skeleton.**
    - Create `01-TOOLS/` directory.
    - Copy `assets/_TEMPLATE/` to `01-TOOLS/_TEMPLATE/`. This template is **generic boilerplate with placeholders (`<NOMBRE_TOOL>`, `<TOOL>_API_KEY`)**. The user copies it manually when adding a tool NOT in the catalog. The skill itself does NOT use `_TEMPLATE/` to generate the detected tools — those come from `providers.yaml`.
@@ -198,7 +162,7 @@ Execute in this exact order. Each step writes to disk; abort and report on first
 4. **Migrate legacy `XX-*` folders.**
    - For each TOOLING file: move to its mapped destination in `01-TOOLS/<X>/`. If the destination file already exists from step 3, the legacy file goes to `01-TOOLS/<X>/migrated/<original-name>` so nothing is overwritten. The user resolves manually.
    - For each DOCS file: move to `02-DOCS/raw/migrated/<original-folder>/<path>`.
-   - For each AMBIGUOUS file: leave in place. Record in the verification report.
+   - For each AMBIGUOUS file: leave in place. Record in the verification report. Never force-classify — a file moved to the wrong place is harder to recover than one left where the user put it.
 5. **Verify migration.**
    - Count files moved vs files originally present. They must match (moved + ambiguous-remaining = original).
    - If counts don't match, abort and report. Don't proceed to deletion.
@@ -216,7 +180,14 @@ Execute in this exact order. Each step writes to disk; abort and report on first
      - Each `01-TOOLS/<TOOL>/README.md` and `CREDENTIALS.md`.
      - Every file under `02-DOCS/raw/migrated/` (from legacy `XX-*` migration in step 4).
      - Root `CLAUDE.md` and `AGENTS.md`.
-   - Use the templates in `references/wiki-*.md` verbatim. Do NOT invent a different format.
+   - Use these templates verbatim; `wiki-protocol.md` is the source of truth for `02-DOCS`, so do NOT invent a different structure or format:
+     - `references/wiki-raw-template.md` — `raw/<topic>/*.md`.
+     - `references/wiki-article-template.md` — `wiki/<topic>/*.md` (OKF v0.1 frontmatter + relative markdown links + `## Related`).
+     - `references/wiki-index-template.md` — `wiki/index.md` (machine catalog; the `.base` views are the human navigation).
+     - `references/wiki-gaps-template.md` — `wiki/gaps.md` (Knowledge Gaps log).
+     - `references/wiki-dashboard-template.html` — the live wiki dashboard, regenerated by Maintenance Pass.
+     - `references/wiki-archive-template.html` — archived query answers (point-in-time, never edited).
+     - `references/wiki-deep-improve-report-template.html` — Deep Improve run reports.
 
 ### Phase 5 — VERIFY
 
@@ -304,20 +275,13 @@ Opt out of the size nudge with `.rsc/.no-claudemd-check`.
 
 ## Invariants
 
-These are not style preferences you can weigh against convenience — each one exists because
-breaking it destroys something the user cannot get back. Violating any aborts the run.
-
-1. **Audit before action.** Nothing is written before the AUDIT is rendered and CONSENT is given.
-2. **Explicit consent only.** Silence, ambiguity, "ok" all mean NO. Only the exact-form strings count as yes.
-3. **No `.env` real, ever.** The skill writes `.env.example` only. Real credentials are the user's responsibility.
-4. **No speculative tools.** A tool is created if and only if the detector found evidence in the user's code. No "we should probably have a Sentry tool too".
-5. **No overwrite without merge.** Existing `CLAUDE.md`, `AGENTS.md`, or tool files are merged additively. The skill never deletes user content.
-6. **Destructive ops require a second consent quoting the path.** "Yes, delete `00-TOOLS`" is different from a generic "yes, proceed".
-7. **Ambiguous files preserve the legacy folder.** Don't force-classify. If you can't classify with confidence, leave it.
-8. **Idempotent.** Running the skill twice produces no extra side effects. Re-scanning a project already canonical detects "nothing to do".
-9. **Out-of-scope dirs are invisible.** `node_modules/`, `.venv/`, `.next/`, `__pycache__/`, `.git/`, `dist/`, `build/`, `.dart_tool/` are never read for detection and never touched.
-10. **`references/wiki-protocol.md` is the source of truth for `02-DOCS`.** Do not invent a different wiki structure.
-11. **Subproject internals are out of scope.** `.env.example`, `requirements.txt`, `package.json`, source files inside subprojects are READ for detection only. They are NEVER moved, renamed, modified, or deleted. The skill operates exclusively on workspace-root artifacts (`CLAUDE.md`, `AGENTS.md`, `01-TOOLS/`, `02-DOCS/`, and `XX-*` legacy folders at the root level).
+The consent, merge and `.env` rules live with the phases that enforce them above. These four are
+scope rules that no single phase owns, and breaking one destroys something the user cannot get back:
+
+1. **No speculative tools.** A tool is created if and only if the detector found evidence in the user's code. No "we should probably have a Sentry tool too".
+2. **Idempotent.** Running the skill twice produces no extra side effects. Re-scanning a project already canonical detects "nothing to do".
+3. **Out-of-scope dirs are invisible.** `node_modules/`, `.venv/`, `.next/`, `__pycache__/`, `.git/`, `dist/`, `build/`, `.dart_tool/` are never read for detection and never touched.
+4. **Subproject internals are out of scope.** `.env.example`, `requirements.txt`, `package.json`, source files inside subprojects are READ for detection only. They are NEVER moved, renamed, modified, or deleted. The skill operates exclusively on workspace-root artifacts (`CLAUDE.md`, `AGENTS.md`, `01-TOOLS/`, `02-DOCS/`, and `XX-*` legacy folders at the root level).
 
 ## Red flags — abort and re-plan
 
@@ -327,30 +291,7 @@ If any of these occur, stop and report:
 - The AUDIT shows zero detected tools AND zero legacy folders AND `CLAUDE.md`/`AGENTS.md` already exist → there's nothing for the skill to do. Tell the user.
 - The user types anything ambiguous after AUDIT → do not infer consent.
 - Migration verification (step 5 of APPLY) shows file count mismatch → abort, don't delete anything.
-- The catalog has no entry for a provider obviously present in code → tell the user, suggest adding a catalog entry, don't fake one.
-
-## References
-
-- `references/providers.yaml` — full provider catalog with detectors and per-file content. Edit this to add a new provider; don't edit `SKILL.md`.
-- `references/claude-md-template.md` — root `CLAUDE.md` template.
-- `references/agents-md-template.md` — root `AGENTS.md` template.
-- `references/tools-readme-template.md` — `01-TOOLS/README.md` catalog template.
-- `references/audit-report-template.md` — text summary format for the in-conversation audit summary.
-- `references/audit-report-template.html` — HTML format for the full per-run audit artifact written to `02-DOCS/audits/`.
-- `references/wiki-protocol.md` — embedded protocol for the `02-DOCS/` chaos→knowledge layer (initialization, **Inbox Sweep**, **Worklog Sweep**, ingest, query, lint, **Continuous Improvement**: Maintenance Pass, Micro-Improve, Deep Improve; Obsidian-native conventions).
-- `references/ingest-formats.md` — multiformat Fetch: how any input (PDF, image, CSV/Excel, JSON/API, html, docx, email, unknown binary) becomes `raw/` markdown with the original preserved in `_originals/`.
-- `references/inbox-readme-template.md` — the `inbox/README.md` drop-zone contract shown to the user.
-- `references/wiki-raw-template.md` — format for `02-DOCS/raw/<topic>/*.md`.
-- `references/wiki-worklog-template.md` — format for `02-DOCS/raw/worklog/*.md` (the work-driven capture; the session as a raw source).
-- `references/wiki-article-template.md` — format for `02-DOCS/wiki/<topic>/*.md` (OKF v0.1 frontmatter + relative markdown links + `## Related`).
-- `references/wiki-index-template.md` — format for `02-DOCS/wiki/index.md` (machine catalog; the `.base` views are the human navigation).
-- `references/obsidian-scaffolding.md` — the vault scaffolding: `.base` views, `.gitignore`, `attachments/`, `.obsidian` config (no vector DB / RAG).
-- `references/daily-curation-automation.md` — portable scheduled compounding pass; on Claude Code wired via the `schedule` skill.
-- `references/wiki-archive-template.html` — HTML format for archived query answers (point-in-time, never edited).
-- `references/wiki-dashboard-template.html` — HTML format for the live wiki dashboard, regenerated by Maintenance Pass.
-- `references/wiki-deep-improve-report-template.html` — HTML format for Deep Improve run reports.
-- `references/wiki-gaps-template.md` — format for `02-DOCS/wiki/gaps.md` (Knowledge Gaps log).
-- `assets/_TEMPLATE/` — the boilerplate copied into every new tool.
+- The catalog has no entry for a provider obviously present in code → tell the user, suggest adding an entry to `references/providers.yaml` (that file, never `SKILL.md`, is where providers are added), don't fake one.
 
 This skill is fully self-contained. No external sub-skill required.
 

package/skills/harness/evals/cases.yaml

@@ -28,7 +28,7 @@ should_trigger:
 should_not_trigger:
   - prompt: "I want to add a single Sentry tool folder to 01-TOOLS. Just copy the template and wire it up."
     route_to: "none"
-    why: "SKILL.md 'Do NOT use' is explicit: adding a single tool is a manual `cp -r 01-TOOLS/_TEMPLATE 01-TOOLS/<X>` — the full SCAN/AUDIT/APPLY harness is overkill, no skill needed."
+    why: "SKILL.md 'Out of scope' is explicit: adding a single tool is a manual `cp -r 01-TOOLS/_TEMPLATE 01-TOOLS/<X>` — the full SCAN/AUDIT/APPLY harness is overkill, no skill needed."
 
   - prompt: "Refactor the runtime database layer in my FastAPI service to use async SQLAlchemy sessions."
     route_to: "fastapi"