@rune-kit/rune 2.22.2 → 2.24.0

package/README.md

@@ -83,7 +83,33 @@ _Methodology: Claude Code CLI headless mode (`claude -p --output-format json`),
 
 ---
 
-## What's New (v2.22.2 — Convergence, Dogfooded)
+## Level Up: Rune Pro
+
+<p align="center">
+  <a href="https://github.com/rune-kit/rune-pro"><img src="assets/pro-upgrade.svg" alt="Rune Pro — Autopilot + Context Intelligence · $49 lifetime" width="100%"></a>
+</p>
+
+Free gives Claude discipline. **Pro** gives it two things the free tier can't:
+
+- **Autopilot** — approve a plan once and walk away. It executes each phase autonomously (self-reviews with zero-HIGH gates, regression-checks against baselines, commits per phase) and **resumes across sessions**. `cook` gets the job done; `autopilot` gets it done while you sleep.
+- **Context Intelligence** — a live **context cockpit** in your statusline (real %, the active skill's mesh layer, cost, predictive "~4m until auto-compaction") that also makes Claude *itself* aware of the pressure, so it saves decisions + progress to `.rune/` **before** compact wipes the session. Zero data loss.
+
+Plus **9 domain packs** (product, sales, data-science, support, growth, media, personal-brand, ecommerce, vietnam). **$49 lifetime, one-time — no subscription.**
+
+<p align="center">
+  <a href="https://github.com/rune-kit/rune-pro"><img src="https://img.shields.io/badge/Unlock_Pro_%2449-lifetime-a855f7?style=for-the-badge" alt="Unlock Rune Pro"></a>
+  <a href="https://rune-kit.github.io/rune#pricing"><img src="https://img.shields.io/badge/See_Pricing-rune--kit.github.io-blue?style=for-the-badge" alt="Pricing"></a>
+</p>
+
+---
+
+## What's New (v2.24.0 — Market Refresh)
+
+> **v2.24.0 (2026-07-11):** Model lineup + platform adapters refreshed to the mid-2026 landscape. Claude tiers across the mesh → **Opus 4.8 / Sonnet 5 / Haiku 4.5**; the **Codex** adapter → **GPT-5.6** `sol` / `terra` / `luna` (verified against codex-cli 0.144.1), with a suggested tier→`model_reasoning_effort` mapping in the generated `AGENTS.md`. **Fable 5** is positioned as an API-only frontier **oracle** target rather than a routing tier (opus stays the ceiling). **Windsurf → Devin Desktop** rebrand documented (emission kept on `.windsurf/`, which Devin still reads). New **goal-first advisory** in `plan`/`ba` to leverage native `/goal` + Managed Agents Outcomes. CI 1572/1572.
+
+> **v2.23.0 (2026-07-04):** Seven platform adapters move to the **Agent Skills open standard** (dir-per-skill `SKILL.md`, discovered and lazy-loaded by each platform's native loader). The headline fix: **Codex** dropped `.codex/skills/` from its scan list, so compiled skills were only findable via the AGENTS.md pointer — agents kept "re-finding" the path mid-session. Codex now emits to **`.agents/skills/`** (scanned CWD → repo root). Same treatment across the fleet: **cursor** `.cursor/rules/*.mdc` → `.cursor/skills/` (Cursor 2.4+ Skills, on-demand instead of always-on), **windsurf** → `.windsurf/skills/` (Cascade Skills), **copilot** → `.github/skills/`, **qoder** → `.qoder/skills/`, and **gemini/qwen** drop their all-skills-always-on context bombs (GEMINI.md bundle, QWEN.md `@import` wall) for native `.gemini/skills/` / `.qwen/skills/` + slim pointer files — a big context-window win on those platforms. Runtime hooks intentionally stay on `.cursor/rules` / `.windsurf/rules` (always-on is correct for hook context). Also fixes a YAML double-escaping bug that corrupted compiled frontmatter for skills with quoted descriptions on 6 platforms, and a duplicate `scripts/` copy in dir-per-skill builds. If you previously built for Codex/Cursor/Windsurf/Copilot/Gemini/Qwen/Qoder: re-run `npx @rune-kit/rune build` and delete the old output dirs. CI 1571/1571.
+
+### Previous (v2.22.2 — Convergence, Dogfooded)
 
 > **v2.22.2 (2026-07-04):** Patch: `rune setup` now installs tier skills into the Claude Code PLUGIN CACHE (newest version dir) instead of the executing package's own root — npx runs were copying Pro skills into npx's ephemeral cache (invisible to the plugin runtime, 'Unknown skill: rune:autopilot' returned), and source-checkout runs polluted the git tree. **v2.22.1:** tier-hook loader accepts PreCompact + SessionStart lifecycle events — required by Pro hooks v1.2.0 (context-reset). **v2.22.0:** The v2.21.0 gates went through a live-fire dogfood: fresh executor agents (zero author context) ran `converge` and `verification` Level 3.5 against a fixture with a dead Submit button, a handler-less Export button, a navigation-anchor decoy, and a declared placeholder. **Both gates caught the dead button with file:line evidence and zero false positives** — and the 16 ambiguities the executors reported became spec fixes: converge v0.2.0 adds a `deferred-debt` class (declared design debt can no longer force a false escalation), a `Plan Claims vs Reality` section (tasks marked `[x]` whose code doesn't exist — surfaced first-class), and derived story verdicts; verification v0.7.0 gets FAIL-dominates precedence, per-route reverse checks, and server/static entry-point exemptions. The dogfood fixture itself shipped as the seed of **`npm run eval`** — a behavioral eval harness that runs a fresh headless agent against fixture repos and asserts outcomes, because structural validation can't prove a skill makes an agent behave. Pro `autopilot` v1.6.0 now explicitly runs Phase 6.5 CONVERGE in autonomous mode.
 
@@ -134,7 +160,7 @@ _Methodology: Claude Code CLI headless mode (`claude -p --output-format json`),
 - **`adversary` v0.2.0 — Mode: oracle** — when `agent.stuck` fires from `debug` (3 disproved hypotheses) or `fix` (2+ failed attempts), oracle-mode dispatches a stateless second-model pass with explicit "no prior context" framing. Bundle format is regex-validated (`[SYSTEM]` invariant role-priming + `[USER]` template + `### File N`), token-capped (100k bundle, 4k per file, 12 files max), citation-required reply contract. Secrets auto-redacted. Breaks the confirmation-bias loop that scout's zoom-out (structural pivot) cannot.
 - **`session-bridge` v0.8.0 — Detach Mode** — async escalation primitive. Heavy-model second-opinion calls (1-10 min wall time) no longer block the primary agent. `.rune/oracle-pending/<sessionId>.json` is the rendezvous file; idempotent dispatch (bundleHash-keyed); 10min default timeout; 24h orphan cleanup on session start. `cook` Phase 4 and `team` Phase 3 reattach via filesystem poll between adjacent tasks.
 - **`context-engine` v1.1.0 — Mode: preview** — pre-flight cost gate. Caller emits `context.preview` BEFORE bundle build with file list + estimated tokens (chars × 0.25). Per-caller thresholds: adversary 50k/100k, team 30k/80k (per worker), review 40k/100k, audit 60k/120k. Action enum `proceed | warn | block`. Override via `RUNE_CONTEXT_THRESHOLDS_<CALLER>`. Stops `team` parallel workstreams from silently blowing $20 of Opus tokens.
-- **Cross-provider model mapping** — 5 non-Anthropic adapters now translate `model: opus|sonnet|haiku` to provider-correct names. **codex** → gpt-5-pro / gpt-5 / gpt-5-mini. **antigravity** → gemini-3-pro / gemini-3-flash / gemini-3-flash-lite. **opencode / openclaw / generic** → tier:heavy / tier:mid / tier:light (provider-agnostic). claude / cursor / windsurf remain no-op (Anthropic backend understands native names).
+- **Cross-provider model mapping** — 5 non-Anthropic adapters now translate `model: opus|sonnet|haiku` to provider-correct names. **codex** → gpt-5.6-sol / gpt-5.6-terra / gpt-5.6-luna. **antigravity** → gemini-3-pro / gemini-3-flash / gemini-3-flash-lite. **opencode / openclaw / generic** → tier:heavy / tier:mid / tier:light (provider-agnostic). claude / cursor / windsurf remain no-op (Anthropic backend understands native names).
 - **Routing clarity sweep** — all 63 SKILL.md descriptions now double-quoted (YAML safety). 13 ambiguous-name skills got explicit "Use when…" routing hints so skill-router doesn't have to guess: ba, completion-gate, constraint-check, doc-processor, integrity-check, logic-guardian, onboard, preflight, sentinel-env, watchdog, worktree, hallucination-guard, mcp-builder.
 - **4 new mesh signals** — `oracle.dispatched`, `oracle.response`, `oracle.failed`, `context.preview`. All registered in Signal Catalog with full emit/listen mapping. `agent.stuck` listeners updated to include adversary in addition to scout.
 - **1,331 tests** — +71 from v2.14.0 across 5 new test files: adapter-model-mapping (18), oracle-bundle-format (19), oracle-pending-schema (16), context-preview-signal (13), skill-description-quality (5).
@@ -591,14 +617,15 @@ Domain-specific skills that plug into the core mesh:
 
 > *Free Rune makes Claude disciplined. Pro makes Claude self-aware.*
 
-**Context Intelligence** — the headline Pro feature. Claude Code auto-compacts at random, wiping your session context. With Pro, Claude **knows** when context is filling up — and proactively saves decisions, progress, and discoveries before compact hits.
+**Context Intelligence** — the headline Pro feature. Claude Code auto-compacts without warning, wiping your session. Pro turns the statusline into a live **context cockpit** and makes Claude *know* when the window is filling — so it saves decisions, progress, and discoveries before compact hits.
 
 ```
-Free:  Claude is blind to context pressure → auto-compact → amnesia
-Pro:   Claude sees real-time context % → saves state → compact → reloads → zero loss
+◈ Opus 4.8 │ ████████░░░ 73% WARM │ ⬡ cook·L1 │ $0.91 │ 5h ██░░ 22%  7d ███░ 61%
 ```
 
-How it works: `rune-pulse` (statusline) reads `context_window.used_percentage` from Claude Code, writes to temp file. `context-inject` hook injects warnings into Claude's context at 70% / 80% / 90%. Claude triggers `session-bridge` + `neural-memory` to persist everything. After `/compact`, `session-start` reloads `.rune/` state — Claude picks up exactly where it left off.
+Three things a blind meter can't do: a **glitch-proof %** (no phantom 0%, no false "100% URGENT" during light work), a **burn-rate ETA** (*"~4m until auto-compaction"*), and **what's eating your context** (*"files 61% · mcp 22%"* → trim the biggest, don't nuke blind). At 70/80/90% real, `context-inject` writes the pressure into Claude's own context → it triggers `session-bridge` + `neural-memory`, you `/compact`, and `session-start` reloads `.rune/`. Zero loss.
+
+Free ships the blind tool-call counter; the real-% cockpit + predictive intelligence is Pro.
 
 **Autopilot** — approve a plan, walk away. Autonomous multi-session execution with zero-HIGH-tolerance quality gates, baseline regression checks, cross-phase coherence review, and structured completion reports. `cook` gets the job done. `autopilot` gets it done while you sleep.
 

package/compiler/__tests__/adapter-model-mapping.test.js

@@ -25,19 +25,19 @@ const baseSkill = {
 describe('codex adapter model mapping', () => {
   const codex = getAdapter('codex');
 
-  test('opus translates to gpt-5-pro', () => {
+  test('opus translates to gpt-5.6-sol', () => {
     const header = codex.generateHeader({ ...baseSkill, model: 'opus' });
-    assert.match(header, /model: gpt-5-pro/);
+    assert.match(header, /model: gpt-5\.6-sol/);
   });
 
-  test('sonnet translates to gpt-5', () => {
+  test('sonnet translates to gpt-5.6-terra', () => {
     const header = codex.generateHeader({ ...baseSkill, model: 'sonnet' });
-    assert.match(header, /model: gpt-5(?!-)/);
+    assert.match(header, /model: gpt-5\.6-terra/);
   });
 
-  test('haiku translates to gpt-5-mini', () => {
+  test('haiku translates to gpt-5.6-luna', () => {
     const header = codex.generateHeader({ ...baseSkill, model: 'haiku' });
-    assert.match(header, /model: gpt-5-mini/);
+    assert.match(header, /model: gpt-5\.6-luna/);
   });
 
   test('omitted model produces no model line', () => {

package/compiler/__tests__/adapters.test.js

@@ -29,7 +29,7 @@ test('listPlatforms returns all 13 platform adapters', () => {
 test('getAdapter returns adapter by name', () => {
   const cursor = getAdapter('cursor');
   assert.strictEqual(cursor.name, 'cursor');
-  assert.strictEqual(cursor.fileExtension, '.mdc');
+  assert.strictEqual(cursor.fileExtension, '.md');
 });
 
 test('getAdapter throws for unknown adapter', () => {
@@ -112,18 +112,45 @@ for (const adapterName of ADAPTER_NAMES) {
   });
 }
 
+// --- Frontmatter escaping (regression: descriptions containing quoted phrases) ---
+
+test('generateHeader escapes quoted descriptions exactly once (valid YAML)', () => {
+  // Parser hands adapters CLEAN text (interior \" already unescaped) — the
+  // adapter re-escapes exactly once. Double-escaping produced `\\"` which
+  // prematurely terminates the YAML scalar.
+  const skill = { name: 'docs', layer: 'L2', group: 'workflow', description: 'The "docs are never outdated" skill' };
+  for (const name of ['cursor', 'windsurf', 'copilot', 'codex', 'antigravity', 'opencode', 'gemini', 'qoder', 'qwen']) {
+    const header = getAdapter(name).generateHeader(skill);
+    assert.ok(header.includes('\\"docs are never outdated\\"'), `${name} should escape quotes once`);
+    assert.ok(!header.includes('\\\\"'), `${name} must not double-escape quotes`);
+  }
+});
+
 // --- Cursor-specific ---
 
-test('cursor adapter generates .mdc frontmatter with alwaysApply', () => {
+test('cursor adapter emits Agent Skills (dir-per-skill SKILL.md)', () => {
   const cursor = getAdapter('cursor');
-  const l0Skill = { name: 'router', layer: 'L0', group: 'meta', description: 'Routes tasks' };
-  const l2Skill = { name: 'fix', layer: 'L2', group: 'workflow', description: 'Fix code' };
+  assert.strictEqual(cursor.outputDir, '.cursor/skills');
+  assert.strictEqual(cursor.useSkillDirectories, true);
+  assert.strictEqual(cursor.skillFileName, 'SKILL.md');
+
+  const header = cursor.generateHeader({ name: 'fix', layer: 'L2', group: 'workflow', description: 'Fix code' });
+  assert.ok(header.includes('name: rune-fix'));
+  assert.ok(header.includes('description: "Fix code"'));
+  assert.ok(!header.includes('alwaysApply'), 'skills format has no alwaysApply');
+});
+
+// --- Windsurf-specific ---
 
-  const l0Header = cursor.generateHeader(l0Skill);
-  assert.ok(l0Header.includes('alwaysApply: true'));
+test('windsurf adapter emits Cascade Skills (dir-per-skill SKILL.md)', () => {
+  const windsurf = getAdapter('windsurf');
+  assert.strictEqual(windsurf.outputDir, '.windsurf/skills');
+  assert.strictEqual(windsurf.useSkillDirectories, true);
+  assert.strictEqual(windsurf.skillFileName, 'SKILL.md');
 
-  const l2Header = cursor.generateHeader(l2Skill);
-  assert.ok(l2Header.includes('alwaysApply: false'));
+  const header = windsurf.generateHeader({ name: 'fix', layer: 'L2', group: 'workflow', description: 'Fix code' });
+  assert.ok(header.includes('name: rune-fix'));
+  assert.ok(header.includes('description: "Fix code"'));
 });
 
 // --- Codex-specific ---
@@ -134,28 +161,50 @@ test('codex adapter uses skill directories', () => {
   assert.strictEqual(codex.skillFileName, 'SKILL.md');
 });
 
+test('codex AGENTS.md documents the tier→model + reasoning-effort mapping', async () => {
+  const codex = getAdapter('codex');
+  const extras = await codex.generateExtraFiles({
+    stats: { skillCount: 65, crossRefsResolved: 204, packCount: 14, files: [] },
+  });
+  const agents = extras.find((e) => e.path === 'AGENTS.md');
+  assert.ok(agents, 'codex emits AGENTS.md');
+  // Guards against a MODEL_MAP / REASONING_EFFORT_MAP interpolation swap.
+  assert.match(agents.content, /opus → `gpt-5\.6-sol` \(`model_reasoning_effort = "high"`\)/);
+  assert.match(agents.content, /sonnet → `gpt-5\.6-terra` \(`model_reasoning_effort = "medium"`\)/);
+  assert.match(agents.content, /haiku → `gpt-5\.6-luna` \(`model_reasoning_effort = "low"`\)/);
+});
+
 // --- New v2.18 adapters: shape + generateExtraFiles contract ---
 
-test('qoder adapter targets .qoder/rules and emits AGENTS.md', async () => {
+test('qoder adapter targets .qoder/skills (dir-per-skill) and emits AGENTS.md', async () => {
   const qoder = getAdapter('qoder');
-  assert.strictEqual(qoder.outputDir, '.qoder/rules');
-  assert.strictEqual(qoder.useSkillDirectories, false);
+  assert.strictEqual(qoder.outputDir, '.qoder/skills');
+  assert.strictEqual(qoder.useSkillDirectories, true);
+  assert.strictEqual(qoder.skillFileName, 'SKILL.md');
   assert.strictEqual(typeof qoder.generateExtraFiles, 'function');
   const extras = await qoder.generateExtraFiles({ stats: { skillCount: 5, packCount: 1, files: [] } });
-  assert.ok(extras.some((e) => e.path === 'AGENTS.md' && e.content.includes('Rune')));
+  assert.ok(extras.some((e) => e.path === 'AGENTS.md' && e.content.includes('.qoder/skills')));
 });
 
-test('copilot adapter targets .github/instructions with .instructions.md ext', async () => {
+test('copilot adapter targets .github/skills with SKILL.md format', async () => {
   const copilot = getAdapter('copilot');
-  assert.strictEqual(copilot.outputDir, '.github/instructions');
-  assert.strictEqual(copilot.fileExtension, '.instructions.md');
-  const header = copilot.generateHeader({ name: 'cook', layer: 'L1', group: 'orchestrator', description: 'Test' });
-  // Per docs.github.com Copilot CLI custom-instructions spec, only `applyTo` is a
-  // documented frontmatter field. Description and tier-hint must live in the body.
-  assert.ok(header.includes('applyTo: "**"'));
-  assert.ok(!/^description:/m.test(header.split('---')[1] || ''), 'description must NOT appear in frontmatter');
+  assert.strictEqual(copilot.outputDir, '.github/skills');
+  assert.strictEqual(copilot.useSkillDirectories, true);
+  assert.strictEqual(copilot.skillFileName, 'SKILL.md');
+  const header = copilot.generateHeader({
+    name: 'cook',
+    layer: 'L1',
+    group: 'orchestrator',
+    description: 'Test',
+    model: 'opus',
+  });
+  // Agent Skills spec: name + description frontmatter; tier hint stays a body comment.
+  assert.ok(header.includes('name: rune-cook'));
+  assert.ok(header.includes('description: "Test"'));
+  assert.ok(header.includes('<!-- tier-hint: tier:heavy -->'));
+  assert.ok(!header.includes('applyTo'), 'skills format has no applyTo');
   const extras = await copilot.generateExtraFiles({ stats: { skillCount: 5, packCount: 1, files: [] } });
-  assert.ok(extras.some((e) => e.path === '.github/copilot-instructions.md'));
+  assert.ok(extras.some((e) => e.path === '.github/copilot-instructions.md' && e.content.includes('.github/skills')));
   assert.ok(extras.some((e) => e.path === 'AGENTS.md'));
 });
 
@@ -201,21 +250,30 @@ test('aider adapter emits .aider.conf.yml with read array', async () => {
   assert.ok(!conf.content.includes('aider/rules/index.md'));
 });
 
-test('qwen adapter emits QWEN.md with @import lines', async () => {
+test('qwen adapter targets .qwen/skills and emits slim QWEN.md pointer', async () => {
   const qwen = getAdapter('qwen');
-  assert.strictEqual(qwen.outputDir, 'qwen/skills');
+  assert.strictEqual(qwen.outputDir, '.qwen/skills');
+  assert.strictEqual(qwen.useSkillDirectories, true);
+  assert.strictEqual(qwen.skillFileName, 'SKILL.md');
   const extras = await qwen.generateExtraFiles({
-    stats: { skillCount: 2, packCount: 0, files: ['rune-cook.md', 'rune-fix.md'] },
+    stats: { skillCount: 2, packCount: 0, files: ['rune-cook/SKILL.md', 'rune-fix/SKILL.md'] },
   });
   const qwenMd = extras.find((e) => e.path === 'QWEN.md');
   assert.ok(qwenMd);
-  assert.ok(qwenMd.content.includes('@qwen/skills/rune-cook.md'));
-  assert.ok(qwenMd.content.includes('@qwen/skills/rune-fix.md'));
+  assert.ok(qwenMd.content.includes('.qwen/skills/rune-<name>/SKILL.md'));
+  assert.ok(!qwenMd.content.includes('@qwen/skills/'), 'no @import lines — skills are lazy-loaded natively');
 });
 
-test('gemini adapter declares generateExtraFiles for bundled GEMINI.md', () => {
+test('gemini adapter targets .gemini/skills and emits slim GEMINI.md pointer', async () => {
   const gemini = getAdapter('gemini');
-  assert.strictEqual(gemini.outputDir, 'gemini/skills');
-  assert.strictEqual(typeof gemini.generateExtraFiles, 'function');
-  // Full bundle test happens in pipeline test (needs filesystem) — contract checked here.
+  assert.strictEqual(gemini.outputDir, '.gemini/skills');
+  assert.strictEqual(gemini.useSkillDirectories, true);
+  assert.strictEqual(gemini.skillFileName, 'SKILL.md');
+  const extras = await gemini.generateExtraFiles({
+    stats: { skillCount: 2, packCount: 0, files: ['rune-cook/SKILL.md', 'rune-fix/SKILL.md'] },
+  });
+  const geminiMd = extras.find((e) => e.path === 'GEMINI.md');
+  assert.ok(geminiMd);
+  assert.ok(geminiMd.content.includes('.gemini/skills/rune-<name>/SKILL.md'));
+  assert.ok(!geminiMd.content.includes('## rune-cook'), 'no bundled H2 sections — skills are lazy-loaded natively');
 });

package/compiler/__tests__/pack-split.test.js

@@ -244,7 +244,10 @@ describe('buildAll split pack auto-discovery', () => {
       assert.strictEqual(stats.packCount, 1, 'should build 1 pack');
 
       // Read the compiled pack output
-      const packOutput = await readFile(path.join(outputRoot, adapter.outputDir, 'rune-ext-test-pack.mdc'), 'utf-8');
+      const packOutput = await readFile(
+        path.join(outputRoot, adapter.outputDir, 'rune-ext-test-pack', 'SKILL.md'),
+        'utf-8',
+      );
 
       // Should contain the pack index body
       assert.ok(packOutput.includes('Pack index body'), 'missing pack index body');
@@ -270,7 +273,7 @@ describe('buildAll split pack auto-discovery', () => {
     await buildAll({ runeRoot, outputRoot: tmp, adapter });
 
     try {
-      const packOutput = await readFile(path.join(tmp, adapter.outputDir, 'rune-ext-ai-ml.mdc'), 'utf-8');
+      const packOutput = await readFile(path.join(tmp, adapter.outputDir, 'rune-ext-ai-ml', 'SKILL.md'), 'utf-8');
 
       // Should be significantly longer than just the index body (~99 lines without skills)
       const lineCount = packOutput.split('\n').length;

package/compiler/__tests__/pipeline.test.js

@@ -58,12 +58,12 @@ describe('full pipeline: cook skill', () => {
 });
 
 describe('full pipeline: fix skill', () => {
-  test('cursor: cross-refs transformed to .mdc', () => {
+  test('cursor: cross-refs transformed to skill references', () => {
     const adapter = getAdapter('cursor');
     const result = transformSkill(fixParsed, adapter);
 
     // Fix references cook, debug, test, etc. — at least some should be transformed
-    assert.ok(result.body.includes('.mdc'), 'cursor output should contain .mdc references');
+    assert.ok(result.body.includes('the rune-'), 'cursor output should contain skill references');
   });
 
   test('generic: cross-refs transformed to descriptive text', () => {
@@ -82,15 +82,15 @@ describe('full pipeline: fix skill', () => {
 // --- Output assembly (header + body + footer) ---
 
 describe('output assembly', () => {
-  test('cursor output assembles into valid .mdc file', () => {
+  test('cursor output assembles into valid SKILL.md file', () => {
     const adapter = getAdapter('cursor');
     const result = transformSkill(cookParsed, adapter);
     const assembled = result.header + result.body + result.footer;
 
     // Should start with YAML frontmatter
-    assert.ok(assembled.startsWith('---\n'), 'cursor .mdc should start with YAML frontmatter');
-    assert.ok(assembled.includes('description:'), 'cursor .mdc should have description');
-    assert.ok(assembled.includes('alwaysApply:'), 'cursor .mdc should have alwaysApply');
+    assert.ok(assembled.startsWith('---\n'), 'cursor SKILL.md should start with YAML frontmatter');
+    assert.ok(assembled.includes('name: rune-'), 'cursor SKILL.md should have name');
+    assert.ok(assembled.includes('description:'), 'cursor SKILL.md should have description');
   });
 
   test('codex output has YAML frontmatter with name', () => {

package/compiler/__tests__/scripts-bundling.test.js

@@ -54,7 +54,7 @@ describe('resolveScriptsPath', () => {
 
 describe('adapter scriptsDir', () => {
   test('flat-file adapters return sibling dir pattern', () => {
-    for (const name of ['cursor', 'windsurf', 'antigravity', 'generic', 'openclaw']) {
+    for (const name of ['generic', 'openclaw']) {
       const adapter = getAdapter(name);
       assert.ok(adapter.scriptsDir, `${name} should have scriptsDir`);
       assert.strictEqual(adapter.scriptsDir('slides'), 'rune-slides-scripts', `${name} scriptsDir mismatch`);
@@ -62,7 +62,17 @@ describe('adapter scriptsDir', () => {
   });
 
   test('directory-per-skill adapters return nested dir pattern', () => {
-    for (const name of ['codex', 'opencode']) {
+    for (const name of [
+      'codex',
+      'opencode',
+      'cursor',
+      'windsurf',
+      'copilot',
+      'qoder',
+      'gemini',
+      'qwen',
+      'antigravity',
+    ]) {
       const adapter = getAdapter(name);
       assert.ok(adapter.scriptsDir, `${name} should have scriptsDir`);
       assert.strictEqual(adapter.scriptsDir('slides'), 'rune-slides/scripts', `${name} scriptsDir mismatch`);
@@ -143,7 +153,7 @@ describe('buildAll with scripts', () => {
     return tmp;
   }
 
-  test('cursor: scripts copied to sibling dir, placeholder resolved', async () => {
+  test('cursor: scripts inside skill directory, placeholder resolved', async () => {
     const tmp = await createTempSkillTree();
     try {
       const outputRoot = path.join(tmp, 'out');
@@ -153,19 +163,19 @@ describe('buildAll with scripts', () => {
       // Scripts copied
       assert.ok(stats.scriptsCopied >= 2, `expected 2+ scripts copied, got ${stats.scriptsCopied}`);
 
-      // Scripts dir exists with files
-      const scriptsOut = path.join(outputRoot, adapter.outputDir, 'rune-test-slide-scripts');
+      // Scripts inside skill dir: .cursor/skills/rune-test-slide/scripts/
+      const scriptsOut = path.join(outputRoot, adapter.outputDir, 'rune-test-slide', 'scripts');
       assert.ok(existsSync(path.join(scriptsOut, 'build-deck.js')), 'build-deck.js missing in output');
       assert.ok(existsSync(path.join(scriptsOut, 'helper.py')), 'helper.py missing in output');
 
-      // Placeholder resolved in .mdc output
-      const mdc = await readFile(path.join(outputRoot, adapter.outputDir, 'rune-test-slide.mdc'), 'utf-8');
-      assert.ok(!mdc.includes('{scripts_dir}'), 'placeholder should be resolved');
-      assert.ok(mdc.includes('.cursor/rules/rune-test-slide-scripts/build-deck.js'), 'resolved path missing');
+      // Placeholder resolved in SKILL.md output
+      const md = await readFile(path.join(outputRoot, adapter.outputDir, 'rune-test-slide', 'SKILL.md'), 'utf-8');
+      assert.ok(!md.includes('{scripts_dir}'), 'placeholder should be resolved');
+      assert.ok(md.includes('.cursor/skills/rune-test-slide/scripts/build-deck.js'), 'resolved path missing');
 
       // Plain skill: no scripts dir created
       assert.ok(
-        !existsSync(path.join(outputRoot, adapter.outputDir, 'rune-test-plain-scripts')),
+        !existsSync(path.join(outputRoot, adapter.outputDir, 'rune-test-plain', 'scripts')),
         'plain skill should not have scripts dir',
       );
     } finally {
@@ -182,7 +192,7 @@ describe('buildAll with scripts', () => {
 
       assert.ok(stats.scriptsCopied >= 2);
 
-      // Scripts inside skill dir: .codex/skills/rune-test-slide/scripts/
+      // Scripts inside skill dir: .agents/skills/rune-test-slide/scripts/
       const scriptsOut = path.join(outputRoot, adapter.outputDir, 'rune-test-slide', 'scripts');
       assert.ok(existsSync(path.join(scriptsOut, 'build-deck.js')), 'build-deck.js missing');
       assert.ok(existsSync(path.join(scriptsOut, 'helper.py')), 'helper.py missing');
@@ -190,7 +200,7 @@ describe('buildAll with scripts', () => {
       // Placeholder resolved
       const md = await readFile(path.join(outputRoot, adapter.outputDir, 'rune-test-slide', 'SKILL.md'), 'utf-8');
       assert.ok(!md.includes('{scripts_dir}'), 'placeholder should be resolved');
-      assert.ok(md.includes('.codex/skills/rune-test-slide/scripts/build-deck.js'), 'resolved path missing');
+      assert.ok(md.includes('.agents/skills/rune-test-slide/scripts/build-deck.js'), 'resolved path missing');
     } finally {
       await rm(tmp, { recursive: true, force: true });
     }
@@ -251,7 +261,7 @@ describe('buildAll with scripts', () => {
 
       const original = await readFile(path.join(tmp, 'skills', 'test-slide', 'scripts', 'build-deck.js'), 'utf-8');
       const copied = await readFile(
-        path.join(outputRoot, adapter.outputDir, 'rune-test-slide-scripts', 'build-deck.js'),
+        path.join(outputRoot, adapter.outputDir, 'rune-test-slide', 'scripts', 'build-deck.js'),
         'utf-8',
       );
       assert.strictEqual(copied, original, 'script content should be identical');
@@ -263,19 +273,19 @@ describe('buildAll with scripts', () => {
   test('multiple adapters resolve different paths', async () => {
     const tmp = await createTempSkillTree();
     try {
-      // Cursor (flat)
+      // Cursor (directory)
       const cursorOut = path.join(tmp, 'out-cursor');
       const cursorAdapter = getAdapter('cursor');
       await buildAll({ runeRoot: tmp, outputRoot: cursorOut, adapter: cursorAdapter });
-      const cursorMdc = await readFile(path.join(cursorOut, '.cursor/rules/rune-test-slide.mdc'), 'utf-8');
-      assert.ok(cursorMdc.includes('.cursor/rules/rune-test-slide-scripts/'), 'cursor path');
+      const cursorMd = await readFile(path.join(cursorOut, '.cursor/skills/rune-test-slide/SKILL.md'), 'utf-8');
+      assert.ok(cursorMd.includes('.cursor/skills/rune-test-slide/scripts/'), 'cursor path');
 
-      // Codex (directory)
+      // Codex (directory, shared .agents/skills standard)
       const codexOut = path.join(tmp, 'out-codex');
       const codexAdapter = getAdapter('codex');
       await buildAll({ runeRoot: tmp, outputRoot: codexOut, adapter: codexAdapter });
-      const codexMd = await readFile(path.join(codexOut, '.codex/skills/rune-test-slide/SKILL.md'), 'utf-8');
-      assert.ok(codexMd.includes('.codex/skills/rune-test-slide/scripts/'), 'codex path');
+      const codexMd = await readFile(path.join(codexOut, '.agents/skills/rune-test-slide/SKILL.md'), 'utf-8');
+      assert.ok(codexMd.includes('.agents/skills/rune-test-slide/scripts/'), 'codex path');
     } finally {
       await rm(tmp, { recursive: true, force: true });
     }

package/compiler/__tests__/transformer.test.js

@@ -27,10 +27,10 @@ describe('transformSkill', () => {
     assert.strictEqual(result.footer, '');
   });
 
-  test('Cursor adapter transforms cross-refs to .mdc format', () => {
+  test('Cursor adapter transforms cross-refs to skill references', () => {
     const cursor = getAdapter('cursor');
     const result = transformSkill(mockSkill, cursor);
-    assert.ok(result.body.includes('@rune-cook.mdc'));
+    assert.ok(result.body.includes('the rune-cook skill'));
     assert.ok(!result.body.includes('rune:cook'));
   });
 
@@ -44,11 +44,11 @@ describe('transformSkill', () => {
     assert.ok(result.footer.length > 0, 'footer should not be empty for non-Claude');
   });
 
-  test('Cursor header contains frontmatter with description', () => {
+  test('Cursor header contains SKILL.md frontmatter with name and description', () => {
     const cursor = getAdapter('cursor');
     const result = transformSkill(mockSkill, cursor);
+    assert.ok(result.header.includes('name: rune-'));
     assert.ok(result.header.includes('description:'));
-    assert.ok(result.header.includes('alwaysApply:'));
   });
 
   test('Generic adapter transforms refs to descriptive text', () => {

package/compiler/adapters/antigravity.js

@@ -74,11 +74,7 @@ export default {
   },
 
   scriptsDir(skillName) {
-    return `rune-${skillName}-scripts`;
-  },
-
-  referencesDir(skillName) {
-    return `rune-${skillName}-references`;
+    return `rune-${skillName}/scripts`;
   },
 
   postProcess(content) {

package/compiler/adapters/codex.js

@@ -1,26 +1,45 @@
 /**
  * OpenAI Codex Adapter
  *
- * Emits SKILL.md files into .codex/skills/{name}/ directories.
+ * Emits SKILL.md files into .agents/skills/{name}/ directories.
  * Codex uses the same SKILL.md frontmatter format (name, description)
  * with markdown body — very close to Rune's native format.
  *
  * Codex project context: AGENTS.md (equivalent to CLAUDE.md)
- * Codex skills dir: .codex/skills/
- * Codex skill format: .codex/skills/{name}/SKILL.md
+ * Codex skills dir: .agents/skills/ (scanned from CWD up to repo root)
+ * Codex skill format: .agents/skills/{name}/SKILL.md
  *
- * MODEL TIER MAPPING (v2.15+):
+ * NOTE: Codex dropped .codex/skills/ from its scan list — .codex/ is config
+ * only (config.toml). Repo skills MUST live in .agents/skills/ to be
+ * auto-discovered. @see https://developers.openai.com/codex/skills
+ *
+ * MODEL TIER MAPPING (v2.15+, lineup refreshed 2026-07):
  * Skill frontmatter `model: opus|sonnet|haiku` (Anthropic naming) is
  * translated to Codex/OpenAI provider-correct model names so the field
  * is meaningful in the compiled output. Unknown tier values pass through.
+ *
+ * Codex GPT-5.6 tiers (verified against codex-cli 0.144.1 + developers.openai.com/codex/models):
+ *   sol  = flagship/complex → opus
+ *   terra = balanced everyday → sonnet
+ *   luna = fast/affordable → haiku
+ * Codex also exposes `model_reasoning_effort = minimal|low|medium|high|xhigh`
+ * (a config.toml key, not per-skill frontmatter) — Rune surfaces the tier→effort
+ * suggestion in AGENTS.md rather than emitting a speculative per-skill field.
  */
 
 import { BRANDING_FOOTER } from '../transforms/branding.js';
 
 const MODEL_MAP = {
-  opus: 'gpt-5-pro',
-  sonnet: 'gpt-5',
-  haiku: 'gpt-5-mini',
+  opus: 'gpt-5.6-sol',
+  sonnet: 'gpt-5.6-terra',
+  haiku: 'gpt-5.6-luna',
+};
+
+// Suggested config.toml `model_reasoning_effort` per tier (global key, documented in AGENTS.md).
+const REASONING_EFFORT_MAP = {
+  opus: 'high',
+  sonnet: 'medium',
+  haiku: 'low',
 };
 
 const TOOL_MAP = {
@@ -37,12 +56,12 @@ const TOOL_MAP = {
 
 export default {
   name: 'codex',
-  outputDir: '.codex/skills',
+  outputDir: '.agents/skills',
   fileExtension: '.md',
   skillPrefix: 'rune-',
   skillSuffix: '',
 
-  // Codex uses directory-per-skill: .codex/skills/{name}/SKILL.md
+  // Codex uses directory-per-skill: .agents/skills/{name}/SKILL.md
   useSkillDirectories: true,
   skillFileName: 'SKILL.md',
 
@@ -105,7 +124,17 @@ export default {
       '',
       '## Skills Directory',
       '',
-      'Skills are located in: .codex/skills/',
+      'Skills are located in: .agents/skills/ (auto-discovered by Codex)',
+      '',
+      '## Model Tiers',
+      '',
+      'Rune skills carry a tier hint (`opus`/`sonnet`/`haiku`). Suggested Codex mapping:',
+      '',
+      `- opus → \`${MODEL_MAP.opus}\` (\`model_reasoning_effort = "${REASONING_EFFORT_MAP.opus}"\`) — architecture / security / planning`,
+      `- sonnet → \`${MODEL_MAP.sonnet}\` (\`model_reasoning_effort = "${REASONING_EFFORT_MAP.sonnet}"\`) — code / edit / review`,
+      `- haiku → \`${MODEL_MAP.haiku}\` (\`model_reasoning_effort = "${REASONING_EFFORT_MAP.haiku}"\`) — scan / read-only`,
+      '',
+      'Set `model` + `model_reasoning_effort` in `~/.codex/config.toml` to match the tier of the task.',
       '',
       '---',
       '> Rune Skill Mesh — https://github.com/rune-kit/rune',