aidevops 2.76.1 → 2.77.1

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.
package/README.md CHANGED
@@ -91,9 +91,10 @@ The result: AI agents that work *with* your development process, not around it.
91
91
 
92
92
  ### Agent Structure
93
93
 
94
- - ~15 main agents (Plan+, Build+, SEO, WordPress, etc.)
95
- - 230+ subagent markdown files organized by domain
96
- - 130+ helper scripts in `.agent/scripts/`
94
+ - 18 primary agents (Plan+, Build+, SEO, WordPress, etc.)
95
+ - 530+ subagent markdown files organized by domain
96
+ - 139+ helper scripts in `.agent/scripts/`
97
+ - 41 slash commands for common workflows
97
98
 
98
99
  <!-- AI-CONTEXT-END -->
99
100
 
@@ -483,7 +484,7 @@ aidevops implements proven agent design patterns identified by [Lance Martin (La
483
484
 
484
485
  | Pattern | Description | aidevops Implementation |
485
486
  |---------|-------------|------------------------|
486
- | **Give Agents a Computer** | Filesystem + shell for persistent context | `~/.aidevops/.agent-workspace/`, 130+ helper scripts |
487
+ | **Give Agents a Computer** | Filesystem + shell for persistent context | `~/.aidevops/.agent-workspace/`, 139+ helper scripts |
487
488
  | **Multi-Layer Action Space** | Few tools, push actions to computer | Per-agent MCP filtering (~12-20 tools each) |
488
489
  | **Progressive Disclosure** | Load context on-demand | Subagent routing with content summaries, YAML frontmatter, read-on-demand |
489
490
  | **Offload Context** | Write results to filesystem | `.agent-workspace/work/[project]/` for persistence |
@@ -681,7 +682,7 @@ The setup script offers to install these tools automatically.
681
682
  | [Outscraper](https://outscraper.com/) | Google Maps & business data extraction | Yes |
682
683
  | [PageSpeed Insights](https://developers.google.com/speed/docs/insights/v5/get-started) | Performance auditing | Yes (Google API) |
683
684
  | [Perplexity](https://docs.perplexity.ai/) | AI-powered research | Yes |
684
- | [Playwright](https://playwright.dev/) | Cross-browser testing | No |
685
+ | [Playwright](https://playwright.dev/) | Cross-browser testing (auto-installed by setup.sh) | No |
685
686
  | [Repomix](https://github.com/yamadashy/repomix) | Codebase packing for AI context | No |
686
687
  | [Serper](https://serper.dev/) | Google Search API (web, images, news) | Yes |
687
688
  | [shadcn/ui](https://ui.shadcn.com/) | UI component library browsing & installation | No |
@@ -700,13 +701,15 @@ The setup script offers to install these tools automatically.
700
701
  - [Context7](https://context7.com/) - Real-time documentation access for thousands of libraries
701
702
  - [Repomix](https://github.com/yamadashy/repomix) - Pack codebases into AI-friendly context
702
703
 
703
- **Browser Automation:**
704
+ **Browser Automation** (6 tools, [benchmarked](#browser-automation)):
704
705
 
705
- - [Stagehand (JavaScript)](https://github.com/browserbase/stagehand) - AI-powered browser automation with natural language
706
- - [Stagehand (Python)](https://github.com/anthropics/stagehand-python) - Python version with Pydantic validation
707
- - [Chrome DevTools](https://chromedevtools.github.io/devtools-protocol/) - Browser automation, performance analysis, debugging
708
- - [Playwright](https://playwright.dev/) - Cross-browser testing and automation
709
- - [Crawl4AI](https://github.com/unclecode/crawl4ai) - Async web crawler optimized for AI
706
+ - [Playwright](https://playwright.dev/) - Fastest engine (0.9s form fill), parallel contexts, extensions, proxy (auto-installed)
707
+ - [dev-browser](https://github.com/nicholasgriffintn/dev-browser) - Persistent profile, stays logged in, ARIA snapshots, pairs with DevTools
708
+ - [agent-browser](https://github.com/vercel-labs/agent-browser) - CLI/CI/CD, `--session` parallel, ref-based element targeting
709
+ - [Crawl4AI](https://github.com/unclecode/crawl4ai) - Bulk extraction, `arun_many` parallel (1.7x), LLM-ready markdown
710
+ - [Playwriter](https://github.com/nicholasgriffintn/playwriter) - Your browser's extensions/passwords/proxy, already unlocked
711
+ - [Stagehand](https://github.com/browserbase/stagehand) - Natural language automation, self-healing selectors
712
+ - [Chrome DevTools MCP](https://github.com/nicholasgriffintn/chrome-devtools-mcp) - Companion: Lighthouse, network throttling, CSS coverage (pairs with any tool)
710
713
  - [Cloudflare Browser Rendering](https://developers.cloudflare.com/browser-rendering/) - Server-side web scraping
711
714
 
712
715
  **SEO & Research:**
@@ -752,6 +755,60 @@ bash .agent/scripts/setup-mcp-integrations.sh stagehand-both # Both versions
752
755
  bash .agent/scripts/setup-mcp-integrations.sh chrome-devtools
753
756
  ```
754
757
 
758
+ ## **Browser Automation**
759
+
760
+ 6 browser tools benchmarked and integrated for AI-assisted web automation, dev testing, and data extraction. Agents automatically select the optimal tool based on task requirements.
761
+
762
+ ### Performance Benchmarks
763
+
764
+ Tested on macOS ARM64, all headless, warm daemon:
765
+
766
+ | Test | Playwright | dev-browser | agent-browser | Crawl4AI | Playwriter | Stagehand |
767
+ |------|-----------|-------------|---------------|----------|------------|-----------|
768
+ | **Navigate + Screenshot** | **1.43s** | 1.39s | 1.90s | 2.78s | 2.95s | 7.72s |
769
+ | **Form Fill** (4 fields) | **0.90s** | 1.34s | 1.37s | N/A | 2.24s | 2.58s |
770
+ | **Data Extraction** (5 items) | 1.33s | **1.08s** | 1.53s | 2.53s | 2.68s | 3.48s |
771
+ | **Multi-step** (click + nav) | **1.49s** | 1.49s | 3.06s | N/A | 4.37s | 4.48s |
772
+ | **Parallel** (3 sessions) | **1.6s** | N/A | 2.0s | 3.0s | N/A | Slow |
773
+
774
+ ### Feature Matrix
775
+
776
+ | Feature | Playwright | dev-browser | agent-browser | Crawl4AI | Playwriter | Stagehand |
777
+ |---------|-----------|-------------|---------------|----------|------------|-----------|
778
+ | **Headless** | Yes | Yes | Yes | Yes | No (your browser) | Yes |
779
+ | **Proxy/VPN** | Full | Via args | No | Full | Your browser | Via args |
780
+ | **Extensions** | Yes (persistent) | Yes (profile) | No | No | Yes (yours) | Possible |
781
+ | **Password managers** | Partial (needs unlock) | Partial | No | No | **Yes** (unlocked) | No |
782
+ | **Parallel sessions** | 5 ctx/2.1s | Shared | 3 sess/2.0s | arun_many 1.7x | Shared | Per-instance |
783
+ | **Session persistence** | storageState | Profile dir | state save/load | user_data_dir | Your browser | Per-instance |
784
+ | **Natural language** | No | No | No | LLM extraction | No | Yes |
785
+ | **Self-healing** | No | No | No | No | No | Yes |
786
+
787
+ ### Tool Selection
788
+
789
+ | Need | Tool | Why |
790
+ |------|------|-----|
791
+ | **Fastest automation** | Playwright | 0.9s form fill, parallel contexts |
792
+ | **Stay logged in** | dev-browser | Profile persists across restarts |
793
+ | **Your extensions/passwords** | Playwriter | Already unlocked in your browser |
794
+ | **Bulk extraction** | Crawl4AI | Purpose-built, parallel, LLM-ready output |
795
+ | **CLI/CI/CD** | agent-browser | No server needed, `--session` isolation |
796
+ | **Unknown pages** | Stagehand | Natural language, self-healing |
797
+ | **Performance debugging** | Chrome DevTools MCP | Companion tool, pairs with any browser |
798
+
799
+ ### AI Page Understanding
800
+
801
+ Agents use lightweight methods instead of expensive vision API calls:
802
+
803
+ | Method | Speed | Token Cost | Use For |
804
+ |--------|-------|-----------|---------|
805
+ | ARIA snapshot | ~0.01s | 50-200 tokens | Forms, navigation, interactive elements |
806
+ | Text extraction | ~0.002s | Text length | Reading content |
807
+ | Element scan | ~0.002s | ~20/element | Form filling, clicking |
808
+ | Screenshot | ~0.05s | ~1K tokens (vision) | Visual debugging only |
809
+
810
+ See [`.agent/tools/browser/browser-automation.md`](.agent/tools/browser/browser-automation.md) for the full decision tree and [`browser-benchmark.md`](.agent/tools/browser/browser-benchmark.md) for reproducible benchmark scripts.
811
+
755
812
  ## **Repomix - AI Context Generation**
756
813
 
757
814
  [Repomix](https://repomix.com/) packages your codebase into AI-friendly formats for sharing with AI assistants. This framework includes optimized Repomix configuration for consistent context generation.
@@ -975,7 +1032,7 @@ aidevops is registered as a **Claude Code plugin marketplace**. Install with two
975
1032
  /plugin install aidevops@aidevops
976
1033
  ```
977
1034
 
978
- This installs the complete framework: 13 domain agents, 225+ subagents, and 130+ helper scripts.
1035
+ This installs the complete framework: 18 primary agents, 530+ subagents, and 139+ helper scripts.
979
1036
 
980
1037
  ### Importing External Skills
981
1038
 
@@ -1055,7 +1112,7 @@ Ordered as they appear in OpenCode Tab selector and other AI assistants (15 tota
1055
1112
 
1056
1113
  ### **Example Subagents with MCP Integration**
1057
1114
 
1058
- These are examples of subagents that have supporting MCPs enabled. See `.agent/` for the full list of 225+ subagents organized by domain.
1115
+ These are examples of subagents that have supporting MCPs enabled. See `.agent/` for the full list of 530+ subagents organized by domain.
1059
1116
 
1060
1117
  | Agent | Purpose | MCPs Enabled |
1061
1118
  |-------|---------|--------------|
@@ -1591,7 +1648,7 @@ bash .agent/scripts/continue-cli.sh review
1591
1648
  **Agent Guides** (in `.agent/`):
1592
1649
 
1593
1650
  - **[API Integrations](.agent/aidevops/api-integrations.md)** - Service APIs
1594
- - **[Browser Automation](.agent/tools/browser/browser-automation.md)** - Web scraping and automation
1651
+ - **[Browser Automation](.agent/tools/browser/browser-automation.md)** - 6 tools benchmarked: decision tree, parallel, extensions, AI understanding
1595
1652
  - **[PageSpeed](.agent/tools/browser/pagespeed.md)** - Performance auditing
1596
1653
  - **[Pandoc](.agent/tools/conversion/pandoc.md)** - Document format conversion
1597
1654
  - **[Security](.agent/aidevops/security.md)** - Enterprise security standards
@@ -1606,8 +1663,8 @@ aidevops/
1606
1663
  ├── AGENTS.md # AI agent guidance (dev)
1607
1664
  ├── .agent/ # Agents and documentation
1608
1665
  │ ├── AGENTS.md # User guide (deployed to ~/.aidevops/agents/)
1609
- │ ├── *.md # 13 main agents
1610
- │ ├── scripts/ # 130+ helper scripts
1666
+ │ ├── *.md # 18 primary agents
1667
+ │ ├── scripts/ # 139+ helper scripts
1611
1668
  │ ├── tools/ # Cross-domain utilities (video, browser, git, etc.)
1612
1669
  │ ├── services/ # External service integrations
1613
1670
  │ └── workflows/ # Development process guides
package/VERSION CHANGED
@@ -1 +1 @@
1
- 2.76.1
1
+ 2.77.1
package/aidevops.sh CHANGED
@@ -3,7 +3,7 @@
3
3
  # AI DevOps Framework CLI
4
4
  # Usage: aidevops <command> [options]
5
5
  #
6
- # Version: 2.76.1
6
+ # Version: 2.77.1
7
7
 
8
8
  set -euo pipefail
9
9
 
package/package.json CHANGED
@@ -1,6 +1,6 @@
1
1
  {
2
2
  "name": "aidevops",
3
- "version": "2.76.1",
3
+ "version": "2.77.1",
4
4
  "description": "AI DevOps Framework - AI-assisted development workflows, code quality, and deployment automation",
5
5
  "type": "module",
6
6
  "main": "index.js",
package/setup.sh CHANGED
@@ -3,7 +3,7 @@
3
3
  # AI Assistant Server Access Framework Setup Script
4
4
  # Helps developers set up the framework for their infrastructure
5
5
  #
6
- # Version: 2.76.1
6
+ # Version: 2.77.1
7
7
  #
8
8
  # Quick Install (one-liner):
9
9
  # bash <(curl -fsSL https://aidevops.dev/install)
@@ -2322,7 +2322,40 @@ setup_browser_tools() {
2322
2322
  print_warning "Node.js not found - Playwriter MCP unavailable"
2323
2323
  fi
2324
2324
 
2325
- print_info "Browser tools: dev-browser (stateful), Playwriter (extension), Stagehand (AI)"
2325
+ # Playwright MCP (cross-browser testing automation)
2326
+ if [[ "$has_node" == "true" ]]; then
2327
+ print_info "Setting up Playwright MCP..."
2328
+
2329
+ # Check if Playwright browsers are installed (--no-install prevents auto-download)
2330
+ if npx --no-install playwright --version &> /dev/null 2>&1; then
2331
+ print_success "Playwright already installed"
2332
+ else
2333
+ local install_playwright
2334
+ read -r -p "Install Playwright MCP with browsers (chromium, firefox, webkit)? (y/n): " install_playwright
2335
+
2336
+ if [[ "$install_playwright" == "y" ]]; then
2337
+ print_info "Installing Playwright browsers..."
2338
+ if npx playwright install; then
2339
+ print_success "Playwright browsers installed"
2340
+ else
2341
+ print_warning "Playwright browser installation failed"
2342
+ print_info "Run manually: npx playwright install"
2343
+ fi
2344
+ else
2345
+ print_info "Skipped Playwright installation"
2346
+ print_info "Install later with: npx playwright install"
2347
+ fi
2348
+ fi
2349
+
2350
+ print_info "Playwright MCP runs via: npx playwright-mcp@latest"
2351
+ fi
2352
+
2353
+ if [[ "$has_node" == "true" ]]; then
2354
+ print_info "Browser tools: dev-browser (stateful), Playwriter (extension), Playwright (testing), Stagehand (AI)"
2355
+ else
2356
+ print_info "Browser tools: dev-browser (stateful), Stagehand (AI)"
2357
+ fi
2358
+ return 0
2326
2359
  }
2327
2360
 
2328
2361
  # Setup AI Orchestration Frameworks (Langflow, CrewAI, AutoGen)