@rune-kit/rune 2.8.0 → 2.10.0

package/agents/watchdog.md

@@ -1,11 +1,26 @@
 ---
 name: watchdog
-description: "Post-deploy monitoring. Checks HTTP status, response times, error detection."
+description: "Post-deploy monitoring — HTTP health checks, response time measurement, error detection. Called by deploy/launch to verify live site. Precision over recall (no false alarms)."
 model: haiku
 subagent_type: general-purpose
 ---
 
-You are the **watchdog** skill of the Rune plugin.
+You are the **watchdog** skill — Rune's deployment health monitor.
 
-Your full skill definition is in `skills/watchdog/SKILL.md` relative to the plugin root.
-Read that file and execute every step precisely. Do not skip steps or improvise beyond what the SKILL.md specifies.
+## Quick Reference
+
+**Workflow:**
+1. Accept base_url + endpoints (default: `/`, `/health`, `/ready`)
+2. HTTP status checks: classify 2xx/3xx/4xx/5xx
+3. Response time: <500ms=FAST, 500-2000ms=ACCEPTABLE, >2000ms=SLOW
+4. Detect errors: 4xx/5xx, timeouts, empty responses
+5. Performance patterns: consistent slowness, cluster degradation, spikes
+6. Report: smoke test results + alerts + performance signals + overall status
+
+**Critical Rules:**
+- Report specific metrics (not vague "seems slow")
+- Include baseline comparison when available
+- Separate perf signals from error alerts (different severity)
+- Precision > recall — no false alarms
+
+Read `skills/watchdog/SKILL.md` for the full specification.

package/agents/worktree.md

@@ -1,11 +1,24 @@
 ---
 name: worktree
-description: "Git worktree lifecycle management."
+description: "Git worktree lifecycle — create isolated workspaces for parallel development, manage branches, cleanup. Max 3 active worktrees. Called by team for parallel streams."
 model: sonnet
 subagent_type: general-purpose
 ---
 
-You are the **worktree** skill of the Rune plugin.
+You are the **worktree** skill — Rune's parallel workspace manager.
 
-Your full skill definition is in `skills/worktree/SKILL.md` relative to the plugin root.
-Read that file and execute every step precisely. Do not skip steps or improvise beyond what the SKILL.md specifies.
+## Quick Reference
+
+**Operations:**
+- **Create:** `git worktree add .claude/worktrees/<name> -b rune/<name>`
+- **List:** active worktrees, verify branch merge status
+- **Cleanup:** remove if merged OR with force (checks uncommitted changes)
+- **Cleanup all:** prune stale branches
+
+**Critical Rules:**
+- MUST use `.claude/worktrees/` directory (never project root)
+- MUST prefix branches with `rune/` (easy cleanup identification)
+- MAX 3 active worktrees enforced
+- MUST check for uncommitted changes before deletion
+
+Read `skills/worktree/SKILL.md` for the full specification.

package/compiler/__tests__/analytics.test.js

@@ -0,0 +1,370 @@
+import assert from 'node:assert/strict';
+import { mkdir, mkdtemp, rm, writeFile } from 'node:fs/promises';
+import os from 'node:os';
+import path from 'node:path';
+import { afterEach, beforeEach, describe, it } from 'node:test';
+import {
+  getAllAnalytics,
+  getModelDistribution,
+  getSessionOverview,
+  getSessionTrend,
+  getSkillChains,
+  getSkillFrequency,
+  getToolDistribution,
+} from '../analytics.js';
+
+// ─── Test Helpers ───
+
+function today() {
+  return new Date().toISOString().slice(0, 10);
+}
+
+function daysAgo(n) {
+  const d = new Date();
+  d.setDate(d.getDate() - n);
+  return d.toISOString().slice(0, 10);
+}
+
+function sessionEntry(overrides = {}) {
+  return {
+    id: `s-${Date.now()}-${Math.random().toString(36).slice(2, 6)}`,
+    date: today(),
+    duration_min: 15,
+    tool_calls: 42,
+    tool_distribution: { Read: 15, Edit: 12, Bash: 8, Grep: 7 },
+    skill_invocations: 5,
+    skills_used: ['cook', 'plan', 'scout'],
+    primary_skill: 'cook',
+    models_used: { opus: 1, sonnet: 2 },
+    skill_durations: { cook: 3000, plan: 1500 },
+    ...overrides,
+  };
+}
+
+function chainEntry(chain, session = 'test-session') {
+  return { session, chain, depth: Array.isArray(chain) ? chain.length : 0 };
+}
+
+async function setupMetrics(tmpDir, sessions = [], chains = [], skillTotals = null) {
+  const metricsDir = path.join(tmpDir, '.rune', 'metrics');
+  await mkdir(metricsDir, { recursive: true });
+
+  if (sessions.length > 0) {
+    const lines = `${sessions.map((s) => JSON.stringify(s)).join('\n')}\n`;
+    await writeFile(path.join(metricsDir, 'sessions.jsonl'), lines);
+  }
+
+  if (chains.length > 0) {
+    const lines = `${chains.map((c) => JSON.stringify(c)).join('\n')}\n`;
+    await writeFile(path.join(metricsDir, 'chains.jsonl'), lines);
+  }
+
+  if (skillTotals) {
+    await writeFile(
+      path.join(metricsDir, 'skills.json'),
+      JSON.stringify({ version: 1, updated: new Date().toISOString(), skills: skillTotals }),
+    );
+  }
+}
+
+// ─── Tests ───
+
+describe('analytics — empty state', () => {
+  let tmpDir;
+  beforeEach(async () => {
+    tmpDir = await mkdtemp(path.join(os.tmpdir(), 'rune-analytics-test-'));
+  });
+  afterEach(async () => {
+    await rm(tmpDir, { recursive: true, force: true });
+  });
+
+  it('returns zeros when no .rune/metrics/ exists', async () => {
+    const overview = await getSessionOverview(tmpDir, 30);
+    assert.equal(overview.total_sessions, 0);
+    assert.equal(overview.avg_duration_min, 0);
+    assert.equal(overview.total_tool_calls, 0);
+  });
+
+  it('returns empty arrays for frequency/distribution', async () => {
+    const skills = await getSkillFrequency(tmpDir, 30);
+    assert.deepEqual(skills, []);
+    const models = await getModelDistribution(tmpDir, 30);
+    assert.deepEqual(models, []);
+    const tools = await getToolDistribution(tmpDir, 30);
+    assert.deepEqual(tools, []);
+    const chains = await getSkillChains(tmpDir, 30);
+    assert.deepEqual(chains, []);
+  });
+
+  it('returns empty trend', async () => {
+    const trend = await getSessionTrend(tmpDir, 30);
+    assert.deepEqual(trend, []);
+  });
+
+  it('getAllAnalytics works with empty data', async () => {
+    const data = await getAllAnalytics(tmpDir, 30);
+    assert.equal(data.overview.total_sessions, 0);
+    assert.ok(data.generated);
+    assert.equal(data.days, 30);
+  });
+});
+
+describe('analytics — with data', () => {
+  let tmpDir;
+  beforeEach(async () => {
+    tmpDir = await mkdtemp(path.join(os.tmpdir(), 'rune-analytics-test-'));
+  });
+  afterEach(async () => {
+    await rm(tmpDir, { recursive: true, force: true });
+  });
+
+  it('getSessionOverview aggregates correctly', async () => {
+    await setupMetrics(tmpDir, [
+      sessionEntry({ duration_min: 10, tool_calls: 30, skill_invocations: 3 }),
+      sessionEntry({ duration_min: 20, tool_calls: 50, skill_invocations: 7 }),
+    ]);
+    const overview = await getSessionOverview(tmpDir, 30);
+    assert.equal(overview.total_sessions, 2);
+    assert.equal(overview.avg_duration_min, 15);
+    assert.equal(overview.total_tool_calls, 80);
+    assert.equal(overview.total_skill_invocations, 10);
+    assert.equal(overview.active_days, 1);
+  });
+
+  it('getSkillFrequency counts sessions per skill', async () => {
+    await setupMetrics(tmpDir, [
+      sessionEntry({ skills_used: ['cook', 'plan'] }),
+      sessionEntry({ skills_used: ['cook', 'debug'] }),
+      sessionEntry({ skills_used: ['plan'] }),
+    ]);
+    const freq = await getSkillFrequency(tmpDir, 30);
+    assert.equal(freq[0].skill, 'cook');
+    assert.equal(freq[0].sessions_count, 2);
+    assert.equal(freq[1].skill, 'plan');
+    assert.equal(freq[1].sessions_count, 2);
+    assert.equal(freq[2].skill, 'debug');
+    assert.equal(freq[2].sessions_count, 1);
+  });
+
+  it('getModelDistribution aggregates model counts', async () => {
+    await setupMetrics(tmpDir, [
+      sessionEntry({ models_used: { opus: 2, sonnet: 3 } }),
+      sessionEntry({ models_used: { sonnet: 1, haiku: 5 } }),
+    ]);
+    const models = await getModelDistribution(tmpDir, 30);
+    const map = Object.fromEntries(models.map((m) => [m.model, m.skill_count]));
+    assert.equal(map.sonnet, 4);
+    assert.equal(map.haiku, 5);
+    assert.equal(map.opus, 2);
+  });
+
+  it('getSessionTrend groups by date', async () => {
+    await setupMetrics(tmpDir, [
+      sessionEntry({ date: daysAgo(1), skill_invocations: 3 }),
+      sessionEntry({ date: daysAgo(1), skill_invocations: 5 }),
+      sessionEntry({ date: today(), skill_invocations: 2 }),
+    ]);
+    const trend = await getSessionTrend(tmpDir, 30);
+    assert.equal(trend.length, 2);
+    assert.equal(trend[0].sessions, 2);
+    assert.equal(trend[0].skill_invocations, 8);
+    assert.equal(trend[1].sessions, 1);
+  });
+
+  it('getSkillChains counts pattern frequency', async () => {
+    await setupMetrics(
+      tmpDir,
+      [sessionEntry({ id: 's1' }), sessionEntry({ id: 's2' })],
+      [
+        chainEntry(['cook', 'plan', 'scout'], 's1'),
+        chainEntry(['cook', 'plan', 'scout'], 's2'),
+        chainEntry(['debug', 'fix'], 's1'),
+      ],
+    );
+    const chains = await getSkillChains(tmpDir, 30);
+    assert.equal(chains[0].chain, 'cook → plan → scout');
+    assert.equal(chains[0].count, 2);
+    assert.equal(chains[1].chain, 'debug → fix');
+    assert.equal(chains[1].count, 1);
+  });
+
+  it('getToolDistribution aggregates tool counts', async () => {
+    await setupMetrics(tmpDir, [
+      sessionEntry({ tool_distribution: { Read: 10, Edit: 5 } }),
+      sessionEntry({ tool_distribution: { Read: 3, Bash: 7 } }),
+    ]);
+    const tools = await getToolDistribution(tmpDir, 30);
+    const map = Object.fromEntries(tools.map((t) => [t.tool, t.count]));
+    assert.equal(map.Read, 13);
+    assert.equal(map.Edit, 5);
+    assert.equal(map.Bash, 7);
+  });
+});
+
+describe('analytics — date filtering', () => {
+  let tmpDir;
+  beforeEach(async () => {
+    tmpDir = await mkdtemp(path.join(os.tmpdir(), 'rune-analytics-test-'));
+  });
+  afterEach(async () => {
+    await rm(tmpDir, { recursive: true, force: true });
+  });
+
+  it('filters sessions by day range', async () => {
+    await setupMetrics(tmpDir, [
+      sessionEntry({ date: today() }),
+      sessionEntry({ date: daysAgo(5) }),
+      sessionEntry({ date: daysAgo(15) }),
+      sessionEntry({ date: daysAgo(45) }),
+    ]);
+    const all = await getSessionOverview(tmpDir, 0);
+    assert.equal(all.total_sessions, 4);
+
+    const week = await getSessionOverview(tmpDir, 7);
+    assert.equal(week.total_sessions, 2);
+
+    const month = await getSessionOverview(tmpDir, 30);
+    assert.equal(month.total_sessions, 3);
+  });
+
+  it('days=0 returns all sessions', async () => {
+    await setupMetrics(tmpDir, [sessionEntry({ date: daysAgo(100) }), sessionEntry({ date: today() })]);
+    const all = await getSessionOverview(tmpDir, 0);
+    assert.equal(all.total_sessions, 2);
+  });
+});
+
+describe('analytics — edge cases', () => {
+  let tmpDir;
+  beforeEach(async () => {
+    tmpDir = await mkdtemp(path.join(os.tmpdir(), 'rune-analytics-test-'));
+  });
+  afterEach(async () => {
+    await rm(tmpDir, { recursive: true, force: true });
+  });
+
+  it('handles malformed JSONL lines gracefully', async () => {
+    const metricsDir = path.join(tmpDir, '.rune', 'metrics');
+    await mkdir(metricsDir, { recursive: true });
+    await writeFile(
+      path.join(metricsDir, 'sessions.jsonl'),
+      `${JSON.stringify(sessionEntry())}\nNOT VALID JSON\n${JSON.stringify(sessionEntry())}\n`,
+    );
+    const overview = await getSessionOverview(tmpDir, 30);
+    assert.equal(overview.total_sessions, 2); // skips malformed line
+  });
+
+  it('handles empty JSONL files', async () => {
+    const metricsDir = path.join(tmpDir, '.rune', 'metrics');
+    await mkdir(metricsDir, { recursive: true });
+    await writeFile(path.join(metricsDir, 'sessions.jsonl'), '');
+    const overview = await getSessionOverview(tmpDir, 30);
+    assert.equal(overview.total_sessions, 0);
+  });
+
+  it('excludes orphaned chains (no matching session)', async () => {
+    await setupMetrics(
+      tmpDir,
+      [sessionEntry({ id: 's1' })],
+      [
+        chainEntry(['cook', 'plan'], 's1'),
+        chainEntry(['debug', 'fix'], 'orphaned-session'), // no matching session
+        chainEntry(null, 's1'), // malformed chain
+      ],
+    );
+    const chains = await getSkillChains(tmpDir, 30);
+    assert.equal(chains.length, 1);
+    assert.equal(chains[0].chain, 'cook → plan');
+  });
+
+  it('handles sessions without optional fields', async () => {
+    await setupMetrics(tmpDir, [
+      { id: 's1', date: today(), duration_min: 5, tool_calls: 10 },
+      // Missing: skill_invocations, skills_used, primary_skill, models_used, skill_durations
+    ]);
+    const overview = await getSessionOverview(tmpDir, 30);
+    assert.equal(overview.total_sessions, 1);
+    const models = await getModelDistribution(tmpDir, 30);
+    assert.deepEqual(models, []);
+  });
+
+  it('handles corrupted skills.json', async () => {
+    const metricsDir = path.join(tmpDir, '.rune', 'metrics');
+    await mkdir(metricsDir, { recursive: true });
+    await writeFile(path.join(metricsDir, 'skills.json'), 'NOT JSON');
+    // Should not throw
+    const data = await getAllAnalytics(tmpDir, 30);
+    assert.ok(data);
+  });
+});
+
+describe('analytics — dashboard HTML generation', () => {
+  it('generates valid HTML from analytics data', async () => {
+    const { generateDashboardHTML } = await import('../dashboard.js');
+
+    const data = {
+      overview: {
+        total_sessions: 5,
+        avg_duration_min: 12,
+        total_tool_calls: 200,
+        total_skill_invocations: 30,
+        active_days: 3,
+      },
+      skillFrequency: [
+        { skill: 'cook', sessions_count: 4 },
+        { skill: 'plan', sessions_count: 2 },
+      ],
+      modelDistribution: [
+        { model: 'sonnet', skill_count: 8 },
+        { model: 'opus', skill_count: 3 },
+      ],
+      sessionTrend: [{ date: '2026-03-30', sessions: 2, duration_min: 20, skill_invocations: 10, tool_calls: 80 }],
+      skillChains: [{ chain: 'cook → plan → scout', count: 3 }],
+      toolDistribution: [
+        { tool: 'Read', count: 50 },
+        { tool: 'Edit', count: 30 },
+      ],
+      skillHeatmap: { heatmap: [], dates: [], maxCount: 1 },
+      sessionTimeline: [],
+      skillMesh: { nodes: [], edges: [], maxCount: 1 },
+      generated: new Date().toISOString(),
+      days: 30,
+    };
+
+    const html = generateDashboardHTML(data);
+    assert.ok(html.includes('<!DOCTYPE html>'));
+    assert.ok(html.includes('Rune Analytics'));
+    assert.ok(html.includes('Skill Mesh'));
+    assert.ok(html.includes('Models'));
+    assert.ok(html.includes('Workflow Chains'));
+    assert.ok(html.includes('Recent Sessions'));
+    assert.ok(html.includes('cook'));
+  });
+
+  it('generates empty state HTML when no data', async () => {
+    const { generateDashboardHTML } = await import('../dashboard.js');
+
+    const data = {
+      overview: {
+        total_sessions: 0,
+        avg_duration_min: 0,
+        total_tool_calls: 0,
+        total_skill_invocations: 0,
+        active_days: 0,
+      },
+      skillFrequency: [],
+      modelDistribution: [],
+      sessionTrend: [],
+      skillChains: [],
+      toolDistribution: [],
+      skillHeatmap: { heatmap: [], dates: [], maxCount: 1 },
+      sessionTimeline: [],
+      skillMesh: { nodes: [], edges: [], maxCount: 1 },
+      generated: new Date().toISOString(),
+      days: 30,
+    };
+
+    const html = generateDashboardHTML(data);
+    assert.ok(html.includes('No analytics data yet'));
+  });
+});

package/compiler/adapters/openclaw.js

@@ -73,7 +73,7 @@ export default {
       name: 'Rune',
       kind: 'skills',
       description:
-        '59-skill mesh for AI coding assistants. Routes all code tasks through specialized skills. 200+ connections, 14 extension packs.',
+        '62-skill mesh for AI coding assistants. Routes all code tasks through specialized skills. 215+ connections, 14 extension packs.',
       version: pluginJson.version || '0.0.0',
       skills: ['./skills'],
       configSchema: {
@@ -116,7 +116,7 @@ export default {
 
 > Less skills. Deeper connections.
 
-**${skills.length}-skill mesh** for AI coding assistants — 5-layer architecture, 200+ connections, 14 extension packs.
+**${skills.length}-skill mesh** for AI coding assistants — 5-layer architecture, 215+ connections, 14 extension packs.
 
 ## Install