feat: perf workflow + enhance hooks/skills

.claude-plugin/marketplace.json

@@ -1,13 +1,26 @@
 {
   "name": "awesome-slash",
-  "description": "8 specialized plugins for AI workflow automation - task orchestration, PR workflow, slop detection, code review, drift detection, enhancement analysis, documentation sync, and repo mapping",
+  "description": "9 specialized plugins for AI workflow automation - task orchestration, PR workflow, slop detection, code review, drift detection, enhancement analysis, documentation sync, repo mapping, and perf investigations",
   "version": "3.3.3",
   "owner": {
     "name": "Avi Fenesh",
     "url": "https://github.com/avifenesh"
   },
   "repository": "https://github.com/avifenesh/awesome-slash",
-  "keywords": ["ai", "llm", "agents", "agentic", "claude-code", "opencode", "codex", "mcp", "automation", "workflow", "code-review", "multi-agent"],
+  "keywords": [
+    "ai",
+    "llm",
+    "agents",
+    "agentic",
+    "claude-code",
+    "opencode",
+    "codex",
+    "mcp",
+    "automation",
+    "workflow",
+    "code-review",
+    "multi-agent"
+  ],
   "plugins": [
     {
       "name": "next-task",
@@ -64,12 +77,29 @@
       "description": "AST-based repository map generation using ast-grep with incremental updates for faster drift analysis",
       "version": "3.3.3",
       "category": "development"
+    },
+    {
+      "name": "perf",
+      "source": "./plugins/perf",
+      "description": "Rigorous performance investigation workflow with baselines, profiling, hypotheses, and evidence-backed decisions",
+      "version": "3.3.3",
+      "category": "development"
     }
   ],
   "mcpServer": {
     "name": "awesome-slash",
     "source": "./mcp-server",
     "description": "Cross-platform MCP server with 3-phase slop detection pipeline and enhance analyzers",
-    "tools": ["workflow_status", "workflow_start", "workflow_resume", "workflow_abort", "task_discover", "review_code", "slop_detect", "enhance_analyze", "repo_map"]
+    "tools": [
+      "workflow_status",
+      "workflow_start",
+      "workflow_resume",
+      "workflow_abort",
+      "task_discover",
+      "review_code",
+      "slop_detect",
+      "enhance_analyze",
+      "repo_map"
+    ]
   }
 }

CHANGELOG.md

@@ -7,16 +7,13 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 
 ## [Unreleased]
 
-## [3.3.3] - 2026-01-29
+### Added
+- **/enhance Hooks Analyzer** - New hook checks for frontmatter completeness and basic safety cues
+- **/enhance Skills Analyzer** - New SKILL.md checks for frontmatter and trigger phrase clarity
+- **Enhance MCP Tool** - `enhance_analyze` now supports `hooks` and `skills` focus targets
 
-### Fixed
-- **Windows Path Template Substitution** - Fixed plugin path template substitution across 18 command/agent files
-  - Replaced `'${CLAUDE_PLUGIN_ROOT}'.replace()` with runtime `process.env` access
-  - Pattern: `(process.env.CLAUDE_PLUGIN_ROOT || process.env.PLUGIN_ROOT || '').replace(/\/g, '/')`
-  - Supports both CLAUDE_PLUGIN_ROOT and PLUGIN_ROOT for cross-platform compatibility
-  - Added validation for missing environment variables in bash sections
-  - Bash pattern: `${CLAUDE_PLUGIN_ROOT:-$PLUGIN_ROOT}` with explicit error handling
-  - Fixes: Template strings in require() paths were not being substituted at runtime on Windows
+### Changed
+- **Enhance Orchestrator** - Expanded to run hooks/skills analyzers alongside existing enhancers
 
 ## [3.3.2] - 2026-01-29
 

README.md

@@ -14,7 +14,7 @@ AI models can write code. That's not the hard part anymore. The hard part is eve
 
 | Section | What's there |
 |---------|--------------|
-| [Commands](#commands) | All 8 commands with jump links |
+| [Commands](#commands) | All 9 commands with jump links |
 | [What This Does](#what-this-project-does) | The problem and how this solves it |
 | [What's Different](#what-makes-this-different) | Why this isn't just another AI tool |
 | [Design Philosophy](#design-philosophy) | The thinking behind the architecture |
@@ -34,8 +34,9 @@ AI models can write code. That's not the hard part anymore. The hard part is eve
 | [`/audit-project`](#audit-project) | Multi-agent code review until issues resolved | [→](#audit-project) |
 | [`/drift-detect`](#drift-detect) | Compares your docs to actual code state | [→](#drift-detect) |
 | [`/repo-map`](#repo-map) | Builds a cached AST repo map for fast analysis | [→](#repo-map) |
-| [`/enhance`](#enhance) | Analyzes prompts, plugins, docs for improvements | [→](#enhance) |
+| [`/enhance`](#enhance) | Analyzes prompts, plugins, agents, docs, hooks, skills | [→](#enhance) |
 | [`/sync-docs`](#sync-docs) | Syncs documentation with code changes | [→](#sync-docs) |
+| [`/perf`](#perf) | Runs structured performance investigations | [→](#perf) |
 
 ---
 
@@ -455,9 +456,9 @@ Tools like `/drift-detect` and planners can use the map instead of re-scanning t
 
 ### /enhance
 
-**Purpose:** Analyzes your prompts, plugins, agents, and docs for improvement opportunities.
+**Purpose:** Analyzes your prompts, plugins, agents, docs, hooks, and skills for improvement opportunities.
 
-**Five analyzers run in parallel:**
+**Seven analyzers run in parallel:**
 
 | Analyzer | What it checks |
 |----------|----------------|
@@ -466,6 +467,8 @@ Tools like `/drift-detect` and planners can use the map instead of re-scanning t
 | claudemd-enhancer | CLAUDE.md/AGENTS.md structure, token efficiency |
 | docs-enhancer | Documentation readability, RAG optimization |
 | prompt-enhancer | Prompt engineering patterns, clarity, examples |
+| hooks-enhancer | Hook frontmatter, structure, safety |
+| skills-enhancer | SKILL.md structure, trigger phrases |
 
 **Each finding includes:**
 - Certainty level (HIGH/MEDIUM/LOW)
@@ -484,6 +487,32 @@ Tools like `/drift-detect` and planners can use the map instead of re-scanning t
 
 ---
 
+### /perf
+
+**Purpose:** Run structured performance investigations with baselines, profiling, and evidence‑backed decisions.
+
+**Usage:**
+
+```bash
+/perf                 # Start new investigation
+/perf --resume        # Resume previous investigation
+```
+
+**Phase flags (advanced):**
+
+```bash
+/perf --phase baseline --command "npm run bench" --version v1.2.0
+/perf --phase breaking-point --command "npm run bench" --param-min 1 --param-max 500
+/perf --phase constraints --command "npm run bench" --cpu 1 --memory 1GB
+/perf --phase hypotheses --hypotheses-file perf-hypotheses.json
+/perf --phase code-paths
+/perf --phase optimization --change "reduce allocations"
+/perf --phase decision --verdict stop --rationale "no measurable improvement"
+/perf --phase consolidation --version v1.2.0
+```
+
+---
+
 ### /sync-docs
 
 **Purpose:** Sync documentation with actual code changes—find outdated refs, update CHANGELOG, flag stale examples.

__tests__/enhance-hooks-skills-analyzer.test.js

@@ -0,0 +1,64 @@
+const fs = require('fs');
+const os = require('os');
+const path = require('path');
+
+const { analyzeHook, analyzeAllHooks } = require('../lib/enhance/hook-analyzer');
+const { analyzeSkill, analyzeAllSkills } = require('../lib/enhance/skill-analyzer');
+
+describe('enhance hook/skill analyzers', () => {
+  let tempDir;
+
+  beforeEach(() => {
+    tempDir = fs.mkdtempSync(path.join(os.tmpdir(), 'enhance-test-'));
+  });
+
+  afterEach(() => {
+    fs.rmSync(tempDir, { recursive: true, force: true });
+  });
+
+  test('analyzeHook detects missing frontmatter', () => {
+    const hookPath = path.join(tempDir, 'hooks');
+    fs.mkdirSync(hookPath, { recursive: true });
+    const filePath = path.join(hookPath, 'pre-commit.md');
+    fs.writeFileSync(filePath, '# Hook\n\nNo frontmatter.');
+
+    const result = analyzeHook(filePath);
+    expect(result.structureIssues.length).toBeGreaterThanOrEqual(1);
+    expect(result.structureIssues.some(issue => /Missing YAML frontmatter/i.test(issue.issue))).toBe(true);
+  });
+
+  test('analyzeAllHooks scans nested hooks directories', () => {
+    const hookA = path.join(tempDir, 'plugins', 'alpha', 'hooks');
+    const hookB = path.join(tempDir, 'tools', 'hooks');
+    fs.mkdirSync(hookA, { recursive: true });
+    fs.mkdirSync(hookB, { recursive: true });
+    fs.writeFileSync(path.join(hookA, 'a.md'), '---\nname: a\ndescription: test\n---\n');
+    fs.writeFileSync(path.join(hookB, 'b.md'), '---\nname: b\ndescription: test\n---\n');
+
+    const results = analyzeAllHooks(tempDir);
+    expect(results.length).toBe(2);
+  });
+
+  test('analyzeSkill detects missing trigger phrase', () => {
+    const skillDir = path.join(tempDir, 'skills', 'example');
+    fs.mkdirSync(skillDir, { recursive: true });
+    const filePath = path.join(skillDir, 'SKILL.md');
+    fs.writeFileSync(filePath, '---\nname: example\ndescription: Helpful skill.\n---\n');
+
+    const result = analyzeSkill(filePath);
+    expect(result.triggerIssues.length).toBe(1);
+    expect(result.triggerIssues[0].issue).toMatch(/trigger phrase/i);
+  });
+
+  test('analyzeAllSkills finds nested SKILL.md files', () => {
+    const skillA = path.join(tempDir, 'skills', 'alpha');
+    const skillB = path.join(tempDir, 'plugins', 'beta', 'skills', 'beta-skill');
+    fs.mkdirSync(skillA, { recursive: true });
+    fs.mkdirSync(skillB, { recursive: true });
+    fs.writeFileSync(path.join(skillA, 'SKILL.md'), '---\nname: alpha\ndescription: Use when user asks about alpha.\n---\n');
+    fs.writeFileSync(path.join(skillB, 'SKILL.md'), '---\nname: beta\ndescription: Use when user asks about beta.\n---\n');
+
+    const results = analyzeAllSkills(tempDir);
+    expect(results.length).toBe(2);
+  });
+});

__tests__/perf-argument-parser.test.js

@@ -0,0 +1,38 @@
+const { parseArguments } = require('../lib/perf/argument-parser');
+
+describe('perf argument parser', () => {
+  it('handles empty input', () => {
+    expect(parseArguments('')).toEqual([]);
+    expect(parseArguments('   ')).toEqual([]);
+    expect(parseArguments(null)).toEqual([]);
+  });
+
+  it('splits basic arguments', () => {
+    expect(parseArguments('--phase baseline --version v1')).toEqual([
+      '--phase',
+      'baseline',
+      '--version',
+      'v1'
+    ]);
+  });
+
+  it('preserves quoted values', () => {
+    const raw = '--command \"npm run bench -- --scenario small\" --quote \"latency spikes\"';
+    expect(parseArguments(raw)).toEqual([
+      '--command',
+      'npm run bench -- --scenario small',
+      '--quote',
+      'latency spikes'
+    ]);
+  });
+
+  it('supports single quotes', () => {
+    const raw = "--quote 'cache miss surge' --phase profiling";
+    expect(parseArguments(raw)).toEqual([
+      '--quote',
+      'cache miss surge',
+      '--phase',
+      'profiling'
+    ]);
+  });
+});

__tests__/perf-baseline.test.js

@@ -0,0 +1,42 @@
+const fs = require('fs');
+const os = require('os');
+const path = require('path');
+
+const stateDir = require('../lib/platform/state-dir');
+const baselineStore = require('../lib/perf/baseline-store');
+const baselineComparator = require('../lib/perf/baseline-comparator');
+
+describe('perf baseline store', () => {
+  let tempDir;
+
+  beforeEach(() => {
+    tempDir = fs.mkdtempSync(path.join(os.tmpdir(), 'perf-baseline-'));
+    process.env.AI_STATE_DIR = '.ai-state';
+    stateDir.clearCache();
+  });
+
+  afterEach(() => {
+    stateDir.clearCache();
+    delete process.env.AI_STATE_DIR;
+    fs.rmSync(tempDir, { recursive: true, force: true });
+  });
+
+  it('writes and reads baselines', () => {
+    baselineStore.writeBaseline('v1.0.0', { metrics: { latency: 120 }, command: 'npm run bench' }, tempDir);
+    const baseline = baselineStore.readBaseline('v1.0.0', tempDir);
+    expect(baseline).not.toBeNull();
+    expect(baseline.metrics.latency).toBe(120);
+  });
+
+  it('rejects invalid baseline versions', () => {
+    expect(() => baselineStore.getBaselinePath('../v1.0.0', tempDir)).toThrow();
+  });
+
+  it('compares baseline metrics', () => {
+    const result = baselineComparator.compareBaselines(
+      { metrics: { latency: 100 } },
+      { metrics: { latency: 125 } }
+    );
+    expect(result.metrics.latency.delta).toBe(25);
+  });
+});

__tests__/perf-benchmark-runner.test.js

@@ -0,0 +1,34 @@
+const { parseMetrics } = require('../lib/perf/benchmark-runner');
+
+describe('perf benchmark parser', () => {
+  it('parses single scenario metrics', () => {
+    const output = [
+      'noise',
+      'PERF_METRICS_START',
+      '{"latency_ms":120,"throughput_rps":450}',
+      'PERF_METRICS_END',
+      'tail'
+    ].join('\n');
+
+    const result = parseMetrics(output);
+    expect(result.ok).toBe(true);
+    expect(result.metrics.latency_ms).toBe(120);
+  });
+
+  it('parses multi-scenario metrics', () => {
+    const output = [
+      'PERF_METRICS_START',
+      '{"scenarios":{"low":{"latency_ms":120},"high":{"latency_ms":450}}}',
+      'PERF_METRICS_END'
+    ].join('\n');
+
+    const result = parseMetrics(output);
+    expect(result.ok).toBe(true);
+    expect(result.metrics.scenarios.low.latency_ms).toBe(120);
+  });
+
+  it('fails when markers are missing', () => {
+    const result = parseMetrics('no metrics here');
+    expect(result.ok).toBe(false);
+  });
+});

__tests__/perf-breaking-point-runner.test.js

@@ -0,0 +1,19 @@
+const { runBreakingPointSearch } = require('../lib/perf/breaking-point-runner');
+
+describe('perf breaking point runner', () => {
+  it('finds breaking point in synthetic range', async () => {
+    const originalEnv = process.env.PERF_PARAM_VALUE;
+
+    const result = await runBreakingPointSearch({
+      command: 'node -e "const v=parseInt(process.env.PERF_PARAM_VALUE||\'0\',10); if(v>=5){process.exit(1);} console.log(\'PERF_METRICS_START\\n{}\\nPERF_METRICS_END\');"',
+      paramEnv: 'PERF_PARAM_VALUE',
+      min: 1,
+      max: 8
+    });
+
+    process.env.PERF_PARAM_VALUE = originalEnv;
+
+    expect(result.attempts).toBeGreaterThan(0);
+    expect(result.breakingPoint).toBe(5);
+  });
+});