feat: quality blitz — accuracy audit all 203 tutorials + 3 new deep tutorials

New tutorials (27,616 total new lines):
- autoresearch (karpathy): 9 files, 3,383 lines — MuonAdamW/Polar Express,
  fixed-budget training loop, val_bpb metric, autonomous ML experiments
- hermes-agent (NousResearch): 9 files, 4,227 lines — 3-layer memory system,
  20+ platform gateways, RL trajectory loop, ACP/MCP ecosystem
- autoagent (HKUDS/MetaChain): 9 files, 4,006 lines — context_variables
  stripping, DockerEnv TCP server, self-modifying pipelines, GAIA eval

Critical rewrites (fabricated/wrong content replaced):
- deer-flow: Airflow-style DAG fiction → real LangGraph StateGraph + 14-stage
  middleware pipeline from bytedance/deer-flow
- anthropic-skills: wrong repo (easing functions) → anthropics/anthropic-quickstarts
- goose: all 8 chapters cited diagnostics-viewer.py → real Rust crate sources

Accuracy audit (200+ tutorials, 1,600+ chapters reviewed):
- Replaced ~500+ wrong source citations (release scripts, CI bots, UI widgets
  cited for every chapter regardless of topic)
- Fixed DSPy API from v1 (dspy-ai/dspy.OpenAI) to v2 (dspy/dspy.LM)
- Fixed Haystack 1.x → 2.x import paths
- Fixed firecrawl org URL (firecrawl → mendableai)
- Fixed outlines org (outlines-dev → dottxt-ai throughout)
- Added phidata→Agno rebranding notice with new repo pointer
- Added ~300+ missing mermaid diagrams across 50+ tutorials
- Added project-archived notices to superagi, sweep
- Fixed star counts: opencode (133k→142k), openspec (35.5k→39.3k)

Patches: openclaw (3 missing subsystems: ACP layer, memory-host-sdk, Plugin SDK)

README: star history chart, back-to-top nav, updated catalog (203 tutorials,
1,624+ chapters), updated discovery table rows

Author: johnxie

.continue/prompts/new-prompt.md

@@ -0,0 +1,7 @@
+---
+name: New prompt
+description: New prompt
+invokable: true
+---
+
+Please write a thorough suite of unit tests for this code, making sure to cover all relevant edge cases

ENHANCEMENT_PLAN.md

@@ -0,0 +1,104 @@
+# Documentation Enhancement Plan
+
+## Current State Analysis
+
+Based on CONTENT_GAPS_ANALYSIS.md:
+- 201 tutorials total
+- 198 with exactly 8 chapters
+- 3 with >8 chapters (n8n-mcp, langchain, ag2)
+- 0 with 0 chapters
+- 0 with partial chapter coverage
+
+## Enhancement Strategy
+
+### Phase 1: High-Traffic Tutorial Regeneration
+**Priority**: Top 10 tutorials by stars from `discoverability/tutorial-source-verification.json`
+
+| Tutorial | Stars | Repo | Status |
+|----------|-------|------|--------|
+| openclaw/openclaw | 341,130 | openclaw/openclaw | Need regeneration |
+| facebook/react | 244,271 | facebook/react | Need regeneration |
+| n8n-io/n8n | 181,679 | n8n-io/n8n | Need regeneration |
+| ollama/ollama | 166,451 | ollama/ollama | Need regeneration |
+| huggingface/transformers | 158,545 | huggingface/transformers | Need regeneration |
+| langflow-ai/langflow | 146,399 | langflow-ai/langflow | Need regeneration |
+| langgenius/dify | 134,981 | langgenius/dify | Need regeneration |
+| anomalyco/opencode | 132,650 | anomalyco/opencode | Need regeneration |
+| langchain-ai/langchain | 131,599 | langchain-ai/langchain | Need regeneration |
+| open-webui/open-webui | 129,246 | open-webui/open-webui | Need regeneration |
+
+### Phase 2: Missing High-Impact Tutorials
+**Priority**: Add tutorials for trending OSS projects not yet covered
+
+**Candidates** (check GitHub for stars > 10K):
+- Vercel AI SDK (22K+ stars) - Already covered
+- Browser Use (85K+ stars) - Already covered  
+- Claude Code (84K+ stars) - Already covered
+- Model Context Protocol servers (82K+ stars) - Already covered
+- Infiniflow RAGFlow (76K+ stars) - Already covered
+- vLLM (74K+ stars) - Already covered
+
+**New additions needed**:
+- Check GitHub for trending repos in AI/agents space
+- Focus on repos with recent activity (pushed_at in last 30 days)
+- Target repos with documentation gaps
+
+### Phase 3: Content Gap Resolution
+**Priority**: Fill missing code examples and depth
+
+**Issues to fix**:
+1. Tutorials with <100 lines in chapters (already addressed in commit 5bda1be)
+2. Missing Mermaid diagrams in architecture chapters
+3. Inconsistent code example quality across tutorials
+4. Missing production deployment examples
+
+### Phase 4: Source Code Extraction Improvements
+**Priority**: Enhance the regeneration script
+
+**Improvements needed**:
+1. Better file prioritization (focus on core modules)
+2. Handle more file types (`.md`, `.json`, `.yaml`, `.toml`)
+3. Better abstraction detection for different languages
+4. Add test file extraction for usage examples
+5. Better Mermaid diagram generation from code structure
+
+## Execution Plan
+
+### Step 1: Regenerate High-Traffic Tutorials
+```bash
+# Run regeneration on top 10 tutorials
+python scripts/regenerate_tutorial_chapters.py \
+  --slugs openclaw,facebook-react,n8n,ollama,huggingface-transformers,langflow,dify,opencode,langchain,open-webui
+```
+
+### Step 2: Add New Tutorials
+1. Identify 5-10 missing high-impact repos
+2. Create tutorial directories with proper structure
+3. Add to `llms.txt` and `llms-full.txt`
+4. Update `discoverability/tutorial-source-verification.json`
+
+### Step 3: Fix Content Gaps
+1. Review tutorials with low chapter counts
+2. Add missing code examples from source repos
+3. Add Mermaid diagrams where missing
+4. Ensure consistent production examples
+
+### Step 4: Improve Source Extraction
+1. Update `regenerate_tutorial_chapters.py`
+2. Add better file filtering logic
+3. Enhance abstraction detection
+4. Add diagram generation from code structure
+
+### Step 5: Quality Verification
+```bash
+# Run health checks
+python scripts/docs_health.py
+```
+
+## Success Metrics
+
+- [ ] All top 10 tutorials have real code examples from source repos
+- [ ] 5-10 new high-impact tutorials added
+- [ ] 0 tutorials with placeholder content
+- [ ] All tutorials pass docs_health.py checks
+- [ ] Source extraction script handles 95%+ of file types

README.md

@@ -1,21 +1,16 @@
+<a name="top"></a>
 <div align="center">
 
-```
-   ___                                         ______          __        ____
-  / _ |_    _____ ___  ___  __ _  ___          / ____/___  ____/ /__     / __ \____  __________
- / __ | |/|/ / -_|_-< / _ \/  ' \/ -_)        / /   / __ \/ __  / _ \   / / / / __ \/ ___/ ___/
-/_/ |_|__,__/\__/___/ \___/_/_/_/\__/        / /___/ /_/ / /_/ /  __/  / /_/ / /_/ / /__(__  )
-                                              \____/\____/\__,_/\___/  /_____/\____/\___/____/
-```
+# Awesome Code Docs
 
-**Deep-dive tutorials for the world's most popular open-source projects**
+**203 deep-dive tutorials for AI agents, LLM frameworks & coding tools**
 
 *Learn how complex systems actually work — not just what they do*
 
 [![Awesome](https://awesome.re/badge.svg)](https://awesome.re)
 [![GitHub stars](https://img.shields.io/github/stars/johnxie/awesome-code-docs?style=social)](https://github.com/johnxie/awesome-code-docs)
-[![Tutorials](https://img.shields.io/badge/tutorials-191-brightgreen.svg)](#-tutorial-catalog)
-[![Sources](https://img.shields.io/badge/source%20repos-191%2F191%20verified-brightgreen.svg)](discoverability/tutorial-source-verification.md)
+[![Tutorials](https://img.shields.io/badge/tutorials-203-brightgreen.svg)](#-tutorial-catalog)
+[![Sources](https://img.shields.io/badge/source%20repos-203%2F203%20verified-brightgreen.svg)](discoverability/tutorial-source-verification.md)
 [![Content Hours](https://img.shields.io/badge/content-2000%2B%20hours-orange.svg)](#-tutorial-catalog)
 [![Last Updated](https://img.shields.io/github/last-commit/johnxie/awesome-code-docs?label=updated)](https://github.com/johnxie/awesome-code-docs/commits/main)
 
@@ -56,6 +51,8 @@ Every tutorial follows a consistent 8-chapter structure:
 
 Each chapter includes **Mermaid architecture diagrams**, **annotated code examples** from the real codebase, and **summary tables** for quick reference.
 
+[![Star History Chart](https://api.star-history.com/svg?repos=johnxie/awesome-code-docs&type=Date)](https://star-history.com/#johnxie/awesome-code-docs&Date)
+
 ---
 
 ## 🔎 Find Tutorials by Goal
@@ -66,14 +63,17 @@ Use this quick-start map if you searched for a specific outcome.
 |:--------------|:-----------|:-----------|
 | open-source vibe coding tools | [Cline](tutorials/cline-tutorial/) | [Roo Code](tutorials/roo-code-tutorial/) → [OpenCode](tutorials/opencode-tutorial/) → [Sweep](tutorials/sweep-tutorial/) → [Tabby](tutorials/tabby-tutorial/) → [Stagewise](tutorials/stagewise-tutorial/) → [bolt.diy](tutorials/bolt-diy-tutorial/) → [VibeSDK](tutorials/vibesdk-tutorial/) → [HAPI](tutorials/hapi-tutorial/) → [Kiro](tutorials/kiro-tutorial/) |
 | spec-driven AI delivery workflows | [OpenSpec](tutorials/openspec-tutorial/) | [Claude Task Master](tutorials/claude-task-master-tutorial/) → [Codex CLI](tutorials/codex-cli-tutorial/) → [OpenCode](tutorials/opencode-tutorial/) → [Kiro](tutorials/kiro-tutorial/) |
-| build AI agents in production | [LangChain](tutorials/langchain-tutorial/) | [LangGraph](tutorials/langgraph-tutorial/) → [CrewAI](tutorials/crewai-tutorial/) → [OpenHands](tutorials/openhands-tutorial/) → [Claude Flow](tutorials/claude-flow-tutorial/) → [Devika](tutorials/devika-tutorial/) → [BabyAGI](tutorials/babyagi-tutorial/) |
+| build AI agents in production | [LangChain](tutorials/langchain-tutorial/) | [LangGraph](tutorials/langgraph-tutorial/) → [CrewAI](tutorials/crewai-tutorial/) → [OpenHands](tutorials/openhands-tutorial/) → [Claude Flow](tutorials/claude-flow-tutorial/) → [Hermes Agent](tutorials/hermes-agent-tutorial/) → [AutoAgent](tutorials/autoagent-tutorial/) → [BabyAGI](tutorials/babyagi-tutorial/) |
 | autonomous AI software engineers | [OpenHands](tutorials/openhands-tutorial/) | [Devika](tutorials/devika-tutorial/) → [SWE-agent](tutorials/swe-agent-tutorial/) → [Aider](tutorials/aider-tutorial/) |
 | task-driven autonomous agents | [BabyAGI](tutorials/babyagi-tutorial/) | [AutoGen](tutorials/autogen-tutorial/) → [CrewAI](tutorials/crewai-tutorial/) → [LangGraph](tutorials/langgraph-tutorial/) |
 | build RAG systems | [LlamaIndex](tutorials/llamaindex-tutorial/) | [Haystack](tutorials/haystack-tutorial/) → [RAGFlow](tutorials/ragflow-tutorial/) |
 | run LLMs locally or at scale | [Ollama](tutorials/ollama-tutorial/) | [llama.cpp](tutorials/llama-cpp-tutorial/) → [vLLM](tutorials/vllm-tutorial/) → [LiteLLM](tutorials/litellm-tutorial/) |
+| autonomous ML training experiments | [autoresearch](tutorials/autoresearch-tutorial/) | [deer-flow](tutorials/deer-flow-tutorial/) → [agno-tutorial](tutorials/agno-tutorial/) |
 | build AI apps with TypeScript/Next.js | [Vercel AI SDK](tutorials/vercel-ai-tutorial/) | [CopilotKit](tutorials/copilotkit-tutorial/) → [LobeChat](tutorials/lobechat-tutorial/) |
 | taskade ai / genesis / mcp workflows | [Taskade](tutorials/taskade-tutorial/) | [Taskade Docs](tutorials/taskade-docs-tutorial/) → [Taskade MCP](tutorials/taskade-mcp-tutorial/) → [Taskade Awesome Vibe Coding](tutorials/taskade-awesome-vibe-coding-tutorial/) → [MCP Servers](tutorials/mcp-servers-tutorial/) |
-| build MCP tools and integrations | [MCP Python SDK](tutorials/mcp-python-sdk-tutorial/) | [FastMCP](tutorials/fastmcp-tutorial/) → [MCP Servers](tutorials/mcp-servers-tutorial/) → [MCP Quickstart Resources](tutorials/mcp-quickstart-resources-tutorial/) → [Create Python Server](tutorials/create-python-server-tutorial/) → [MCP Docs Repo](tutorials/mcp-docs-repo-tutorial/) → [Create TypeScript Server](tutorials/create-typescript-server-tutorial/) → [Awesome MCP Servers](tutorials/awesome-mcp-servers-tutorial/) → [Composio](tutorials/composio-tutorial/) → [Daytona](tutorials/daytona-tutorial/) → [GenAI Toolbox](tutorials/genai-toolbox-tutorial/) → [awslabs/mcp](tutorials/awslabs-mcp-tutorial/) → [MCP Inspector](tutorials/mcp-inspector-tutorial/) → [MCP Registry](tutorials/mcp-registry-tutorial/) → [MCP Specification](tutorials/mcp-specification-tutorial/) → [MCP TypeScript SDK](tutorials/mcp-typescript-sdk-tutorial/) → [MCP Go SDK](tutorials/mcp-go-sdk-tutorial/) → [MCP Rust SDK](tutorials/mcp-rust-sdk-tutorial/) → [MCP Java SDK](tutorials/mcp-java-sdk-tutorial/) → [MCP C# SDK](tutorials/mcp-csharp-sdk-tutorial/) → [MCP Swift SDK](tutorials/mcp-swift-sdk-tutorial/) → [MCP Kotlin SDK](tutorials/mcp-kotlin-sdk-tutorial/) → [MCP Ruby SDK](tutorials/mcp-ruby-sdk-tutorial/) → [MCP PHP SDK](tutorials/mcp-php-sdk-tutorial/) → [MCP Ext Apps](tutorials/mcp-ext-apps-tutorial/) → [MCPB](tutorials/mcpb-tutorial/) → [use-mcp](tutorials/use-mcp-tutorial/) → [MCP Use](tutorials/mcp-use-tutorial/) → [MCP Chrome](tutorials/mcp-chrome-tutorial/) → [Firecrawl MCP Server](tutorials/firecrawl-mcp-server-tutorial/) |
+| build MCP tools and integrations | [MCP Python SDK](tutorials/mcp-python-sdk-tutorial/) | [FastMCP](tutorials/fastmcp-tutorial/) → [MCP Servers](tutorials/mcp-servers-tutorial/) → [Awesome MCP Servers](tutorials/awesome-mcp-servers-tutorial/) → [MCP Inspector](tutorials/mcp-inspector-tutorial/) → [MCP TypeScript SDK](tutorials/mcp-typescript-sdk-tutorial/) → [Composio](tutorials/composio-tutorial/) → [see all 30+ MCP tutorials →](tutorials/mcp-python-sdk-tutorial/) |
+
+<div align="right"><a href="#top">⬆ Back to top</a></div>
 
 ---
 
@@ -93,23 +93,27 @@ Quick jump links:
 - [Search Intent Map](discoverability/search-intent-map.md)
 - [Category Hubs](#category-hubs)
 
+<div align="right"><a href="#top">⬆ Back to top</a></div>
+
 ---
 
 ## ✅ Source Verification Status
 
 All tutorial indexes were re-verified against referenced upstream GitHub repositories on **2026-03-20**:
 
-- tutorials scanned: **191**
-- tutorials with source repos: **191**
+- tutorials scanned: **203**
+- tutorials with source repos: **203**
 - tutorials with unverified source repos: **0**
-- unique verified source repos: **201**
+- unique verified source repos: **203**
 
 Verification artifacts:
 
 - [Tutorial Source Verification Report](discoverability/tutorial-source-verification.md)
 - [Tutorial Source Verification JSON](discoverability/tutorial-source-verification.json)
 - verification script: [`scripts/verify_tutorial_sources.py`](scripts/verify_tutorial_sources.py)
 
+<div align="right"><a href="#top">⬆ Back to top</a></div>
+
 ---
 
 ## 🧬 Taskade Ecosystem Snapshot (Verified 2026-03-21)
@@ -157,7 +161,7 @@ Data source: GitHub REST API (`stargazers_count`, `pushed_at`) via `scripts/refr
 ```
  ╔════════════════════════════════════════════════════════════╗
  ║  🤖  AI & AGENTS  │  🔧  DEV TOOLS  │  🗄️  DATA  │  🎤 SPEECH  ║
- ║   70+ tutorials    │   46 tutorials  │  14 tutorials │  3 tutorials  ║
+ ║   83+ tutorials    │   50+ tutorials │  14 tutorials │  3 tutorials  ║
  ╚════════════════════════════════════════════════════════════╝
 ```
 
@@ -197,7 +201,9 @@ Build autonomous AI systems that reason, plan, and collaborate.
 | **[BabyAGI](tutorials/babyagi-tutorial/)** | 18K+ | Python | Task-driven autonomous agent patterns, memory, and BabyAGI 2o/3 evolution |
 | **[AgenticSeek](tutorials/agenticseek-tutorial/)** | 25.4K+ | Python | Local-first autonomous agent with multi-agent planning, browsing, and coding workflows |
 | **[Agno](tutorials/agno-tutorial/)** | 38.3K+ | Python | Multi-agent systems with memory, orchestration, and AgentOS runtime |
-| **[AutoAgent](tutorials/autoagent-tutorial/)** | 8.6K+ | Python | Zero-code agent creation through natural-language workflows |
+| **[AutoAgent](tutorials/autoagent-tutorial/)** | 9.1K+ | Python | Zero-code agent creation through natural-language workflows and self-developing pipelines |
+| **[autoresearch](tutorials/autoresearch-tutorial/)** | 71K+ | Python | AI agent that autonomously runs ML training experiments overnight, optimizing val_bpb on a single GPU |
+| **[Hermes Agent](tutorials/hermes-agent-tutorial/)** | 66K+ | Python | Self-hosted personal AI successor to OpenClaw — multi-platform, skill learning, RL trajectory generation |
 | **[ADK Python](tutorials/adk-python-tutorial/)** | 18.1K+ | Python | Production-grade agent engineering with Google's Agent Development Kit |
 | **[Qwen-Agent](tutorials/qwen-agent-tutorial/)** | 13.5K+ | Python | Tool-enabled agent framework with MCP, RAG, and multi-modal workflows |
 | **[Strands Agents](tutorials/strands-agents-tutorial/)** | 5.2K+ | Python | Model-driven agents with native MCP, hooks, and deployment patterns |
@@ -458,6 +464,8 @@ Voice recognition, audio processing, and multimodal AI applications.
 | **[Whisper.cpp](tutorials/whisper-cpp-tutorial/)** | 37K+ | C++ | Speech recognition on edge devices |
 | **[OpenAI Realtime Agents](tutorials/openai-realtime-agents-tutorial/)** | 6.7K+ | TypeScript | Voice-first AI agents with WebRTC |
 
+<div align="right"><a href="#top">⬆ Back to top</a></div>
+
 ---
 
 ## 🗺️ Learning Paths
@@ -594,6 +602,8 @@ Dyad ──→ bolt.diy ──→ Stagewise ──→ Cline ──→ Roo Code 
 
 **Duration:** 35-50 hours | **Difficulty:** Intermediate to Advanced
 
+<div align="right"><a href="#top">⬆ Back to top</a></div>
+
 ---
 
 ## 📊 Collection Stats
@@ -602,8 +612,8 @@ Dyad ──→ bolt.diy ──→ Stagewise ──→ Cline ──→ Roo Code 
 ╔══════════════════════════════════════════════════════════╗
 ║                  COLLECTION OVERVIEW                     ║
 ╠══════════════════════════════════════════════════════════╣
-║  📦 Total Tutorials        191                           ║
-║  📝 Numbered Chapters      1,528+                        ║
+║  📦 Total Tutorials        203                           ║
+║  📝 Numbered Chapters      1,624+                        ║
 ║  📏 Tutorial Markdown      1,048,763 lines               ║
 ║  ⏱️  Estimated Hours        2,000+                        ║
 ║  ✅ Local Broken Links      0                             ║
@@ -616,6 +626,8 @@ Stats are synchronized against:
 - `tutorials/tutorial-manifest.json`
 - `scripts/docs_health.py` baseline checks
 
+<div align="right"><a href="#top">⬆ Back to top</a></div>
+
 ---
 
 ## 🛠️ How Tutorials Are Built
@@ -642,6 +654,8 @@ Inspired by [Tutorial-Codebase-Knowledge](https://github.com/The-Pocket/Tutorial
 | **[Claude Code](https://claude.ai)** | Codebase analysis and tutorial writing |
 | **[GitHub Pages](https://pages.github.com)** | Tutorial hosting with Jekyll |
 
+<div align="right"><a href="#top">⬆ Back to top</a></div>
+
 ---
 
 ## 🤝 Contributing
@@ -670,6 +684,8 @@ We welcome contributions! Here's how you can help:
 
 **[Open an Issue](https://github.com/johnxie/awesome-code-docs/issues/new)** to suggest a new tutorial or report a problem.
 
+<div align="right"><a href="#top">⬆ Back to top</a></div>
+
 ---
 
 ## 🌍 Community

discoverability/query-hub.md

@@ -2,7 +2,7 @@
 
 Auto-generated high-intent query landing surface mapped to the most relevant tutorials.
 
-- Total tutorials indexed: **201**
+- Total tutorials indexed: **203**
 - Query hubs: **6**
 - Source: `scripts/generate_discoverability_assets.py`