@defai.digital/automatosx 12.7.0 → 12.8.2

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
Files changed (4) hide show
  1. package/README.md +74 -85
  2. package/dist/index.js +12796 -8906
  3. package/dist/mcp/index.js +2981 -157
  4. package/package.json +12 -4
package/README.md CHANGED
@@ -12,7 +12,7 @@
12
12
  [![Ubuntu](https://img.shields.io/badge/Ubuntu-24.04-blue.svg)](https://ubuntu.com)
13
13
  [![License](https://img.shields.io/badge/license-Apache--2.0-yellow.svg)](LICENSE)
14
14
 
15
- **Status**: ✅ **Production Ready** | v12.7.0 | Cognitive Prompt Engineering Framework
15
+ **Status**: ✅ **Production Ready** | v12.8.2 | CI/CD Improvements
16
16
 
17
17
  > 🎯 **What AutomatosX Does**: Adds 20+ specialized agents, persistent memory, workflow automation, and 80% cost savings to Claude Code/Codex - **without changing how you work**.
18
18
 
@@ -48,6 +48,9 @@
48
48
  ✅ Multi-Provider: Auto-routes to Claude/Gemini/OpenAI optimally
49
49
  ✅ Complete Observability: JSONL trace logs for every decision
50
50
  ✅ Workflow Automation: YAML specs for complex multi-step tasks
51
+ ✅ Autonomous Bug Fixing: Detect and fix timer leaks, resource issues (v12.6.0)
52
+ ✅ Autonomous Refactoring: Improve code quality with safety guards (v12.7.0)
53
+ ✅ AST-Based Detection: Reduced false positives via TypeScript AST analysis (v12.8.0)
51
54
  ```
52
55
 
53
56
  **Total Monthly Cost**: $60/mo (same unlimited usage)
@@ -116,7 +119,7 @@ ax setup --claude-code
116
119
  **What you get**:
117
120
  - ✅ 18 slash commands (`/agent-backend`, `/agent-frontend`, etc.)
118
121
  - ✅ Multi-agent orchestration (`/automatosx`)
119
- - ✅ MCP server with 17 tools
122
+ - ✅ MCP server with 27 tools (including bugfix and refactor)
120
123
  - ✅ Auto-generated manifests
121
124
  - ✅ One-command diagnostics (`ax doctor --claude-code`)
122
125
 
@@ -153,66 +156,76 @@ ax setup
153
156
 
154
157
  ---
155
158
 
156
- ## 💬 Natural Language Examples (How You'll Use It)
159
+ ## 💬 Just Talk Naturally (Interactive Mode)
157
160
 
158
- ### Example 1: Multi-Agent Collaboration
161
+ **Simply ask Claude Code to work with AutomatosX agents. No special commands needed.**
162
+
163
+ ### 🐛 Find and Fix Bugs
159
164
 
160
- **You say to Claude Code**:
161
165
  ```
162
- "Work with AutomatosX agents to build a payment system.
163
- Let the backend agent implement the API, security agent audit it,
164
- and quality agent write tests. Save everything to memory."
166
+ "Please work with ax to find and fix bugs in src/core/"
165
167
  ```
166
168
 
167
- **What AutomatosX does automatically**:
168
- 1. Product agent designs architecture
169
- 2. Backend agent implements API (routed to Gemini - cost-efficient)
170
- 3. Security agent audits code (routed to Claude - best reasoning)
171
- 4. Quality agent writes tests (routed to Gemini - fast)
172
- 5. Everything saved to memory for future tasks
173
- 6. Complete trace log for observability
169
+ AutomatosX automatically:
170
+ - Scans for timer leaks, resource issues, type errors
171
+ - Shows you what it found with severity levels
172
+ - Fixes auto-fixable bugs with verification
173
+ - Creates backups and rolls back if tests fail
174
+
175
+ ### 📝 Review and Improve PRD
176
+
177
+ ```
178
+ "Please review the PRD in docs/spec.md and discuss with ax agents to improve it"
179
+ ```
174
180
 
175
- **Cost**: $0 additional (using CLI subscriptions)
176
- **Time**: 60% faster (parallel execution)
177
- **Quality**: 3 specialized agents vs 1 generalist
181
+ AutomatosX automatically:
182
+ - Product agent analyzes requirements completeness
183
+ - Architecture agent identifies technical gaps
184
+ - Security agent flags compliance concerns
185
+ - Returns actionable suggestions for improvement
178
186
 
179
- ### Example 2: Persistent Memory
187
+ ### 🔧 Refactor Code
180
188
 
181
- **Day 1 - You say**:
182
189
  ```
183
- "Design a calculator app with add/subtract features.
184
- Save the architecture to AutomatosX memory."
190
+ "Please refactor the session manager using ax, focus on reducing complexity"
185
191
  ```
186
192
 
187
- **Day 7 - You say**:
193
+ AutomatosX automatically:
194
+ - Measures current metrics (complexity, duplication)
195
+ - Identifies refactoring opportunities
196
+ - Applies changes with safety guards
197
+ - Verifies metrics improved (rollback if not)
198
+
199
+ ### 🔒 Security Audit
200
+
188
201
  ```
189
- "Implement the calculator API we designed last week."
202
+ "Ask the ax security agent to audit the authentication module"
190
203
  ```
191
204
 
192
- **AutomatosX automatically**:
193
- - Searches memory for "calculator" design
194
- - Injects full context into prompt (< 1ms search)
195
- - Zero token waste (no repetition)
196
- - Perfect continuity (like talking to same person)
205
+ AutomatosX automatically:
206
+ - Routes to Claude (best reasoning for security)
207
+ - Checks for OWASP vulnerabilities
208
+ - Reviews auth flow and token handling
209
+ - Saves findings to memory for future reference
197
210
 
198
- **Savings**: 50-80% fewer tokens (no context repetition)
211
+ ### 💾 Remember Anything
199
212
 
200
- ### Example 3: Cost Optimization
213
+ ```
214
+ "Save this API design to ax memory for later"
215
+ ```
201
216
 
202
- **You say**:
203
217
  ```
204
- "Review this 10,000-line codebase for bugs.
205
- Use cost-efficient routing - this is routine analysis."
218
+ "What did we decide about the database schema last week?"
206
219
  ```
207
220
 
208
- **AutomatosX automatically**:
209
- - Routes to Gemini (routine task = cost-efficient)
210
- - Uses Gemini Advanced unlimited plan ($20/mo)
211
- - Instead of Claude API ($150-300/mo for this volume)
212
- - Same quality, 80% cost savings
221
+ AutomatosX automatically:
222
+ - Stores context with full-text search (< 1ms)
223
+ - Retrieves relevant history when needed
224
+ - Zero token waste on repetition
225
+
226
+ ---
213
227
 
214
- **API Cost (without AutomatosX)**: ~$45 (10K lines × $15/1M tokens)
215
- **CLI Cost (with AutomatosX)**: $0 (already paying $20/mo Gemini Advanced)
228
+ > 📚 **Want CLI commands?** See [Full Documentation](docs/full-features.md) for `ax bugfix`, `ax refactor`, and all CLI options.
216
229
 
217
230
  ---
218
231
 
@@ -312,57 +325,31 @@ steps:
312
325
 
313
326
  **AutomatosX executes everything automatically**, with optimal provider routing.
314
327
 
315
- ### 6. Autonomous Bug Fixing (v12.4.0+)
316
-
317
- **Automatically detect and fix common bugs in your codebase:**
328
+ ### 6. Autonomous Bug Fixing (v12.6.0)
318
329
 
319
- ```bash
320
- # Scan and fix timer leaks, missing destroy(), and resource issues
321
- ax bugfix
330
+ **Just ask naturally:**
331
+ ```
332
+ "Please work with ax to find and fix bugs in this project"
333
+ ```
322
334
 
323
- # Dry-run mode (preview fixes without applying)
324
- ax bugfix --dry-run
335
+ **What it detects**: Timer leaks, missing destroy(), promise leaks, event leaks, race conditions, memory leaks, uncaught promises, deprecated APIs, security issues, type errors (13 types total)
325
336
 
326
- # Focus on specific directory
327
- ax bugfix --scope src/core/
337
+ **Safety**: Backups created, typecheck verification, test verification, automatic rollback on failure
328
338
 
329
- # Fix specific bug types only
330
- ax bugfix --types timer_leak,missing_destroy
331
- ```
339
+ > 📚 **Full CLI reference**: See [Bugfix Documentation](docs/guides/bugfix.md) for `ax bugfix --dry-run`, `--staged`, `--json`, pre-commit hooks, and ignore comments
332
340
 
333
- **What AutomatosX detects and fixes**:
334
- - **Timer Leaks**: `setInterval`/`setTimeout` without cleanup
335
- - **Missing Destroy**: EventEmitters without `destroy()` method
336
- - **Promise Timeout Leaks**: Timeouts not cleared in error paths
337
- - **Event Leaks**: Event listeners without cleanup
341
+ ### 7. Autonomous Code Refactoring (v12.7.0)
338
342
 
339
- **Safe by default**:
340
- - Creates backups before any changes
341
- - Runs TypeScript typecheck after each fix
342
- - Runs tests to verify fixes don't break functionality
343
- - Automatic rollback if verification fails
344
-
345
- **Example output**:
343
+ **Just ask naturally:**
344
+ ```
345
+ "Please refactor src/core/ using ax, focus on reducing complexity"
346
346
  ```
347
- [ax bugfix] Scanning src/ for bugs...
348
- [ax bugfix] Found 18 potential issues:
349
- - 12 timer_leak (setInterval without .unref())
350
- - 4 missing_destroy (EventEmitter subclass)
351
- - 2 promise_timeout_leak
352
347
 
353
- [ax bugfix] Fixing timer_leak in src/core/worker-pool.ts:447
354
- - Replaced setInterval with createSafeInterval
355
- - Verification: TypeScript ✓ Tests ✓
348
+ **What it improves**: Dead code removal, type safety (`any` types), complex conditionals, hardcoded values, naming, duplication, readability, performance
356
349
 
357
- [ax bugfix] Complete: 15/18 bugs fixed, 3 require manual review
358
- ```
350
+ **Safety**: Metrics must improve or changes are rejected. Overengineering guards prevent unnecessary abstractions. Automatic rollback on failure.
359
351
 
360
- **MCP Integration** - Use bugfix via Quality agent from Claude Code:
361
- ```
362
- # Via Quality agent (recommended - full context and orchestration)
363
- run_agent({ agent: "quality", task: "scan src/ for code health issues" })
364
- run_agent({ agent: "quality", task: "run bugfix workflow with dry-run" })
365
- ```
352
+ > 📚 **Full CLI reference**: See [Refactor Documentation](docs/guides/refactor.md) for `ax refactor scan`, `--dry-run`, `--no-llm`, and all focus areas
366
353
 
367
354
  ---
368
355
 
@@ -661,6 +648,8 @@ Save everything to shared memory."
661
648
  - [Persistent Memory Guide](docs/guides/memory.md) - Context management
662
649
  - [Workflow Automation](docs/guides/spec-kit-guide.md) - YAML workflows
663
650
  - [Observability](docs/advanced/observability.md) - Trace logging
651
+ - [Autonomous Bug Fixing](docs/guides/bugfix.md) - Detect and fix bugs automatically
652
+ - [Autonomous Refactoring](docs/guides/refactor.md) - Improve code quality with safety guards
664
653
 
665
654
  ### For Advanced Users (5% Who Need CLI)
666
655
  - **[Full Features List](docs/full-features.md)** - All CLI commands and capabilities
@@ -680,7 +669,7 @@ Save everything to shared memory."
680
669
 
681
670
  ```bash
682
671
  npm install -g @defai.digital/automatosx
683
- ax --version # v12.0.1
672
+ ax --version # v12.7.0
684
673
  ```
685
674
 
686
675
  ### Initialize Your Project
@@ -714,8 +703,8 @@ ax setup
714
703
 
715
704
  ## 🚦 Production Readiness
716
705
 
717
- ✅ **v12.0.1 Released** - Bug Fixes & Provider Architecture Improvements
718
- ✅ **2,512+ Tests Passing** - Comprehensive coverage
706
+ ✅ **v12.7.0 Released** - Cognitive Prompt Engineering Framework + Autonomous Refactoring
707
+ ✅ **3,148+ Tests Passing** - Comprehensive coverage
719
708
  ✅ **TypeScript Strict Mode** - Type-safe codebase
720
709
  ✅ **Zero Resource Leaks** - Clean shutdown guaranteed
721
710
  ✅ **Cross-Platform** - macOS, Windows, Ubuntu