@litfamily/litmuse 0.1.1 → 0.1.2

package/plugin/skills/lit-typographic-motion/styles/tidal.json

@@ -0,0 +1,41 @@
+{
+  "schema": "litmuse.motion-style/v1",
+  "id": "tidal",
+  "name": "Tidal",
+  "summary": "Indigo ground with a slow teal-to-violet flow, off-white type that drifts with it, one coral moment.",
+  "light": false,
+  "typePreset": "tidal",
+  "palette": { "ground": "#0E1420", "surface": "#18223A", "ink": "#E8ECEF", "muted": "#A9B3C1", "signal": "#5FB3C4", "accent": "#E07856", "line": "#3A4660", "stopA": "#124559", "stopB": "#4C3B6E" },
+  "accents": [{ "role": "a warm highlight", "color": "#E8C07A" }, { "role": "a second current", "color": "#7FD6B0" }, { "role": "a far layer", "color": "#B9A3E3" }],
+  "textRoles": ["ink", "muted", "signal", "accent", "ground"],
+  "faces": {
+    "display": { "family": "Archivo", "weight": 700, "stretch": "100%" },
+    "headline": { "family": "Archivo", "weight": 700, "stretch": "100%" },
+    "body": { "family": "Pretendard", "weight": 400 },
+    "label": { "family": "MesloLGS NF", "weight": 400 }
+  },
+  "scale": {
+    "16:9": { "display": 116, "headline": 72, "body": 40, "label": 26 },
+    "9:16": { "display": 112, "headline": 76, "body": 46, "label": 32 }
+  },
+  "leading": { "display": 1.12, "headline": 1.22, "body": 1.6, "label": 1.4 },
+  "motion": {
+    "ease": { "enter": [0.37, 0, 0.63, 1], "move": [0.45, 0, 0.55, 1], "exit": [0.5, 0, 0.75, 0] },
+    "dur": { "micro": 0.25, "move": 0.9, "hero": 1.2, "exit": 0.6 },
+    "stagger": { "letter": 0.04, "word": 0.12, "item": 0.14 },
+    "spring": { "micro": { "zeta": 1, "hz": 2.5 }, "panel": { "zeta": 1, "hz": 1.2 }, "headline": { "zeta": 1, "hz": 1.4 }, "subject": { "zeta": 0.9, "hz": 1 }, "camera": { "zeta": 1, "hz": 0.35 } },
+    "secondary": { "delay": 0.2, "travel": 0.6 },
+    "ambient": { "travel": 0.05 },
+    "push": 0.025
+  },
+  "transitions": ["dissolve", "match", "wipe"],
+  "ground": { "kind": "flow", "stops": ["stopA", "stopB"] },
+  "sound": { "timbre": "felt-strings", "key": "D", "mode": "major", "tempo": 84 },
+  "score": { "genre": "deep-house" },
+  "beds": [
+    { "timbre": "felt-strings", "key": "D", "mode": "major", "tempo": 84 },
+    { "timbre": "glass", "key": "F#", "mode": "minor", "tempo": 76 },
+    { "timbre": "soft-keys", "key": "Bb", "mode": "major", "tempo": 72 }
+  ],
+  "bans": ["a rainbow or hue-cycling ramp", "bloom on the flow", "hard snaps on type", "glitch as texture", "flow so fast that held type smears"]
+}

package/plugin/skills/readme-studio/SKILL.md

@@ -47,7 +47,7 @@ Produces a README whose every claim traces to the repository and whose every com
 
    A clean report is structure-only: `source_contents_compared: false`, `badge_truth_checked: false`, `factual_accuracy: "not-checked"`. It rejects unsafe paths, symlinks, missing sources, duplicate `id` values in `claims[]` and malformed badge URLs, and it certifies no fact. Reopen the cited sources before using a claim.
 3. **Run the commands.** Execute install-free checks in the sandbox: `<cli> --help`, `<cli> --version`, the quick-start example, the test command. Keep exit codes and the first lines of output in `commands.txt`. A command that needs network, credentials or a publish step goes into the README with its prerequisites, and the facts file labels it as not run.
-4. **Interview only when material.** Follow `references/production-interview.md`. Ask one high-impact question per turn (audience conflict, content scope, static versus motion delivery that changes accessibility or package contents, unresolved output location) and carry answers forward. A default is fine for details the brief settled, delegated or left immaterial.
+4. **Interview, theme first when building.** Follow `references/production-interview.md`. Ask one high-impact question per turn (audience conflict, content scope, static versus motion delivery that changes accessibility or package contents, unresolved output location) and carry answers forward. A default is fine for details the brief settled, delegated or left immaterial. Exception: on a build or refresh request, the first question is always the cover direction (static cover, motion header, both, neither — recommended first) unless the brief names it or says "you choose": then take the recommended, log a `decision` line, and carry on. Headless runs take the recommended and log it the same way.
 5. **Outline.** Default order:
    1. Title and one-sentence description
    2. Why it exists (two or three sentences, concrete problem)

package/plugin/skills/start-work/SKILL.md

@@ -106,6 +106,12 @@ action named, on the paths named, for this run.
 - **Resume** happens when the user grants the boundary in their own words, or sends a bare `lit` / `lit continue` /
   `/start-work`. Re-read the plan, ledger and evidence before acting; quote the grant in a `decision` line with the
   scope it covers. A grant for one action does not extend to the next boundary.
+- **Typed answers.** A `lit approve` / `lit deny` prompt routes here and carries the boundary's decision: `lit approve`
+  allows the named action on the named paths for this run (record it as the grant), `lit deny` refuses it for this
+  run (carry on without it, and say so if a task cannot be done otherwise). With no paused boundary waiting, either
+  word changes nothing: say so in one sentence and start no other work.
+- **Typed start.** A prompt that names the plan file (`lit start plans/<slug>.md`) is the approval for that plan:
+  begin row 1 without a separate approval turn. Ask only when no single plan can be chosen.
 - **Cancel** when the user says stop. Stop spawning work, cancel running children (`subagent_cancel`), stop the
   processes the run started, remove the run's scratch files, append `CANCELLED: <reason>` as a ledger `note`, leave
   the plan as it stands (ticked rows stay ticked; an open row stays open), set the goal record to `blocked` with the