FROM python:3.12-slim

ARG PORT=8051

ENV PYTHONDONTWRITEBYTECODE=1
ENV PYTHONUNBUFFERED=1

WORKDIR /app

# Install system dependencies
RUN apt-get update && \
    apt-get install -y --no-install-recommends \
        build-essential \
        curl \
        ca-certificates \
        git && \
    rm -rf /var/lib/apt/lists/*

# Install uv tool
RUN pip install --no-cache-dir uv

# Copy the source
COPY . .

# Install project in editable mode and setup
RUN pip install --no-cache-dir --system -e . && \
    crawl4ai-setup

EXPOSE ${PORT}

CMD ["python", "src/crawl4ai_mcp.py"]

# Ignore VCS and common build artifacts
.git
.gitignore
__pycache__/
**/*.pyc
**/*.pyo
*.pyd
*.pyc
env/
ENV/
venv/
ENV3/
*.env
*.venv
build/
dist/
*.egg-info/
*.pytest_cache/
.pytest_cache/
.mypy_cache/
 *.DS_Store
node_modules/
logs/

FROM python:3.12-slim

ARG PORT=8051

ENV PYTHONDONTWRITEBYTECODE=1
ENV PYTHONUNBUFFERED=1

WORKDIR /app

# Install system dependencies
RUN apt-get update && \
    apt-get install -y --no-install-recommends \
        build-essential \
        curl \
        ca-certificates \
        git && \
    rm -rf /var/lib/apt/lists/*

# Install uv tool (if available in environment). Fallback to plain pip for dependencies.
RUN pip install --no-cache-dir uv

# Copy the source
COPY . .

# Install project in editable mode and setup
RUN pip install --no-cache-dir -e . && \
    crawl4ai-setup

EXPOSE ${PORT}

CMD ["python", "src/crawl4ai_mcp.py"]

Build timed out after 600s

FROM python:3.12-slim

ARG PORT=8051

ENV PYTHONDONTWRITEBYTECODE=1
ENV PYTHONUNBUFFERED=1

WORKDIR /app

# Install system dependencies
RUN apt-get update && \
    apt-get install -y --no-install-recommends \
        build-essential \
        curl \
        ca-certificates \
        git && \
    rm -rf /var/lib/apt/lists/*

# Install uv tool (if available in environment). Fallback to plain pip for dependencies.
RUN pip install --no-cache-dir uv

# Copy the source
COPY . .

# Install project dependencies from pyproject.toml in editable mode
RUN pip install --no-cache-dir -e .

EXPOSE ${PORT}

CMD ["python", "src/crawl4ai_mcp.py"]

crawl4ai_mcp.egg-info
__pycache__
.venv
.env

FROM python:3.12-slim

ARG PORT=8051

ENV PYTHONDONTWRITEBYTECODE=1
ENV PYTHONUNBUFFERED=1

WORKDIR /app

# Install system dependencies
RUN apt-get update && \
    apt-get install -y --no-install-recommends \
        build-essential \
        curl \
        ca-certificates \
        git && \
    rm -rf /var/lib/apt/lists/*

# Copy pyproject first to leverage docker cache for dependencies
COPY pyproject.toml ./

# Install project dependencies from pyproject.toml in editable mode
RUN pip install --no-cache-dir -e .

# Copy the rest of the source
COPY . .

EXPOSE ${PORT}

CMD ["python", "src/crawl4ai_mcp.py"]

FROM python:3.12-slim

ARG PORT=8051

ENV PYTHONDONTWRITEBYTECODE=1
ENV PYTHONUNBUFFERED=1

WORKDIR /app

# Copy pyproject first to leverage docker cache for dependencies
COPY pyproject.toml ./

# Install project dependencies from pyproject.toml in editable mode
RUN pip install --no-cache-dir -e .

# Copy the rest of the source
COPY . .

EXPOSE ${PORT}

CMD ["python", "src/crawl4ai_mcp.py"]

Actionable summary (under 1500 chars)

- Purpose: MCP server "mcp-crawl4ai-rag" for web crawling with Crawl4AI, RAG, AI hallucination detection, and Neo4j knowledge-graph repository parsing. Uses Supabase for storage.

- Core components/libraries:
  - FastMCP, AsyncWebCrawler, CrossEncoder, AIScriptAnalyzer, HallucinationReporter
  - Knowledge graph: KnowledgeGraphValidator, DirectNeo4jExtractor
  - Storage: Supabase client
  - Code utilities: generate_code_example_summary, extract_code_blocks, update_source_info, extract_source_summary
  - Path hookup: knowledge_graphs path added to sys.path; dotenv loading from project_root/.env

- Environment/config flags (examples):
  - USE_RERANKING (default false): load CrossEncoder reranker
  - USE_KNOWLEDGE_GRAPH (default false): enable Neo4j components
  - NEO4J_URI, NEO4J_USER, NEO4J_PASSWORD: Neo4j credentials
  - USE_AGENTIC_RAG (default false): enable code-example extraction
  - USE_HYBRID_SEARCH (default false): hybrid vector+keyword search
  - TRANSPORT: 'sse' or stdio for MCP server
  - HOST/PORT, etc.

- Lifespan management:
  - crawl4ai_lifespan creates headless browser (BrowserConfig headless), initializes crawler, Supabase client, optional reranking model, and optional knowledge-graph components (validator and repo_extractor) if enabled.
  - On exit, closes components safely.

- Key tools exposed (examples):
  - crawl_single_page(ctx, url): crawl one page, chunk content, store in Supabase, optionally extract and store code examples
  - smart_crawl_url(ctx, url, max_depth, max_concurrent, chunk_size): auto-detects sitemap/.txt/web page; stores chunks; returns crawl summary
  - get_available_sources(ctx): lists available sources from DB
  - perform_rag_query(ctx, query, source, match_count): vector/hybrid RAG query
  - search_code_examples(ctx, query, source_id, match_count): code example search (if enabled)
  - check_ai_script_hallucinations(ctx, script_path): validate script with knowledge graph
  - query_knowledge_graph(ctx, command): Neo4j knowledge-graph explorer (repos, explore, classes, class, method, query)
  - parse_github_repository(ctx, repo_url): clone/analyze GitHub repo; stores stats in Neo4j
  - parse_* helpers: crawl_markdown_file, crawl_batch, crawl_recursive_internal_links
  - main(): runs SDP or stdio transport based on TRANSPORT

- Knowledge-graph command surface (as in query_knowledge_graph):
  - repos, explore <repo>, classes [<repo>], class <name>, method <name> [<class>], query <cypher>

- Data flow and storage:
  - For crawls: generate chunks via smart_chunk_markdown; gather metadata; update_source_info first; then add_documents_to_supabase
  - Code examples: extracted in parallel (ThreadPoolExecutor) if USE_AGENTIC_RAG=true; stored via add_code_examples_to_supabase
  - Source summaries generated with extract_source_summary and stored

- Notable helpers/logic:
  - is_sitemap, is_txt, parse_sitemap for crawl routing
  - smart_chunk_markdown respects code blocks, paragraphs, and sentences
  - validate_neo4j_connection and format_neo4j_error for user-friendly Neo4j errors

- Error handling (examples):
  - Neo4j: “Neo4j authentication failed. Check NEO4J_USER and NEO4J_PASSWORD.”; “Cannot connect to Neo4j. Check NEO4J_URI and ensure Neo4j is running.”
  - GitHub repo: “Repository URL must start with https:// or git@” and “Repository URL is required”
  - Script path: “Script path is required”; “Cannot read script file: ...”

- Run details:
  - Entry point: asyncio.run(main())
  - Environment-driven behavior predominates (feature flags and Neo4j/Supabase wiring)

FROM python:3.12-slim

ARG PORT=8051

ENV PYTHONDONTWRITEBYTECODE=1
ENV PYTHONUNBUFFERED=1

WORKDIR /app

# Install system dependencies
RUN apt-get update && \
    apt-get install -y --no-install-recommends \
        curl \
        ca-certificates \
        git && \
    rm -rf /var/lib/apt/lists/*

# Install Python dependencies (pinned to known versions from pyproject)
RUN pip install --no-cache-dir \
    crawl4ai==0.6.2 \
    mcp==1.7.1 \
    supabase==2.15.1 \
    openai==1.71.0 \
    dotenv==0.9.9 \
    sentence-transformers>=4.1.0 \
    neo4j>=5.28.1

# Copy project metadata first to enable docker cache for dependencies
COPY pyproject.toml ./

# Copy the rest of the source
COPY . .

EXPOSE ${PORT}

CMD ["python", "src/crawl4ai_mcp.py"]

Build timed out after 600s

FROM python:3.12-slim

ARG PORT=8051

WORKDIR /app

# Install uv helper (used by crawl4ai for installation)
RUN pip install uv

# Copy the MCP server files
COPY . .

# Install project dependencies from source using uv, in editable mode, then run setup if provided
RUN uv pip install -e . && \
    crawl4ai-setup

EXPOSE ${PORT}

CMD ["python", "src/crawl4ai_mcp.py"]

- Exact error message and exit code:
  error: No virtual environment found; run `uv venv` to create an environment, or pass `--system` to install into a non-virtual environment
  Exit code: 2

- Failing command/step:
  [5/5] RUN uv pip install -e . &&     crawl4ai-setup

- Missing packages or files mentioned:
  No virtual environment found (uv expected a venv). You need to create one with `uv venv` or install system-wide with `--system`.

- Version mismatch info (if any):
  - uv-0.11.6 was installed earlier in the build
  - Base image used: python:3.12-slim
  No explicit version mismatch reported beyond the uv virtual env requirement.

FROM python:3.12-slim

ARG PORT=8051

WORKDIR /app

# System dependencies for building and running
RUN apt-get update && \
    apt-get install -y --no-install-recommends \
        build-essential \
        curl \
        ca-certificates \
        git \
    && rm -rf /var/lib/apt/lists/*

# Install Python dependencies directly (system-wide)
RUN pip install --no-cache-dir \
    crawl4ai==0.6.2 \
    mcp==1.7.1 \
    supabase==2.15.1 \
    openai==1.71.0 \
    dotenv==0.9.9 \
    sentence-transformers>=4.1.0 \
    neo4j>=5.28.1

# Copy the MCP server files
COPY . .

EXPOSE ${PORT}

CMD ["python", "src/crawl4ai_mcp.py"]

FROM python:3.12-slim

ARG PORT=8051

ENV PYTHONDONTWRITEBYTECODE=1
ENV PYTHONUNBUFFERED=1

WORKDIR /app

# Install system dependencies needed for builds
RUN apt-get update && \
    apt-get install -y --no-install-recommends \
        build-essential \
        curl \
        ca-certificates \
        git && \
    rm -rf /var/lib/apt/lists/*

# Copy pyproject to leverage Docker cache for dependencies
COPY pyproject.toml ./

# Install Python dependencies (from pyproject) in editable mode
RUN pip install --no-cache-dir -e .

# Copy the rest of the source code
COPY . .

EXPOSE ${PORT}

CMD ["python", "src/crawl4ai_mcp.py"]

FROM python:3.12-slim

ARG PORT=8051

ENV PYTHONDONTWRITEBYTECODE=1
ENV PYTHONUNBUFFERED=1

WORKDIR /app

# Install system dependencies for builds
RUN apt-get update && \
    apt-get install -y --no-install-recommends \
        build-essential \
        curl \
        ca-certificates \
        git \
        libffi-dev \
        libssl-dev && \
    rm -rf /var/lib/apt/lists/*

# Install Python dependencies from PyPI (matching pyproject.toml)
RUN pip install --no-cache-dir \
    crawl4ai==0.6.2 \
    mcp==1.7.1 \
    supabase==2.15.1 \
    openai==1.71.0 \
    dotenv==0.9.9 \
    sentence-transformers>=4.1.0 \
    neo4j>=5.28.1

# Copy the source code
COPY . .

EXPOSE ${PORT}

CMD ["python", "src/crawl4ai_mcp.py"]