itzzjb/DockAI - AI Dockerfile Generation Framework

The Customizable AI Framework that generates, validates, and optimizes production-ready Dockerfiles.

View on GitHub

Trust Signals

Scorecard Score
not yet scored
Maintenance Recency
Activelast commit Apr 1, 2026
License
MIT

Pinned Snippet

workflow.ymlSHA-pinned
uses: itzzjb/dockai@12a147d0460abe40bd15d00068421cc97e61e153 # v4.0.7

tags can be moved; commit SHAs can't. why a SHA?

namedescriptionrequireddefault
llm_providerLLM provider to use (openai, azure, gemini, anthropic, ollama)noopenai
openai_api_keyOpenAI API Key (required for openai provider)no
azure_openai_api_keyAzure OpenAI API Key (required for azure provider)no
azure_openai_endpointAzure OpenAI endpoint URL (required for azure provider)no
azure_openai_api_versionAzure OpenAI API versionno2024-02-15-preview
google_api_keyGoogle API Key (required for gemini provider)no
anthropic_api_keyAnthropic API Key (required for anthropic provider)no
project_pathPath to the project to dockerizeno.
model_analyzerModel for analyzer agent (default: gpt-4o-mini for openai, gemini-1.5-flash for gemini)no""
model_blueprintModel for blueprint agent (Chief Architect)no""
model_generatorModel for generator agent (default: gpt-4o for openai, gemini-1.5-pro for gemini)no""
model_reviewerModel for security reviewer agentno""
model_reflectorModel for failure reflection agentno""
model_error_analyzerModel for error classification agentno""
model_iterative_improverModel for iterative improvement agentno""
max_retriesMax attempts to fix a failing Dockerfileno3
skip_hadolintSkip Hadolint Dockerfile linting (true/false)nofalse
skip_security_scanSkip Trivy security scan (true/false)nofalse
strict_securityFail on any vulnerability (true/false)nofalse
max_image_size_mbMax image size in MB (0 to disable)no500
skip_health_checkSkip health checks during validation (true/false)nofalse
skip_security_reviewSkip AI security review for faster generation (auto-skipped for scripts)nofalse
validation_memoryMemory limit for container validation (e.g., 512m, 1g)no512m
validation_cpusCPU limit for container validationno1.0
validation_pidsMax processes for container validationno100
max_file_charsMax characters to read per file (default: 200000)no200000
max_file_linesMax lines to read per file (default: 5000)no5000
truncation_enabledForce-enable smart truncation for critical files (true/false)nofalse
token_limitToken threshold that auto-enables truncation (default: 50000)no50000
use_ragEnable RAG (Retrieval-Augmented Generation) for intelligent context retrieval (true/false, default: true in v4.0)notrue
embedding_modelHuggingFace embedding model for RAG (default: all-MiniLM-L6-v2)noall-MiniLM-L6-v2
read_all_filesRead all source files for complete context (true/false, default: true)notrue
analyzer_instructionsCustom instructions for the analyzer (project discovery agent)no""
generator_instructionsCustom instructions for the generator (Dockerfile creation agent)no""
blueprint_instructionsCustom instructions for the blueprint agent (Chief Architect)no""
reviewer_instructionsCustom instructions for security review (auditor agent)no""
reflector_instructionsCustom instructions for failure analysis (detective agent)no""
error_analyzer_instructionsCustom instructions for error classification (troubleshooter agent)no""
iterative_improver_instructionsCustom instructions for applying fixes (surgical fix agent)no""
prompt_analyzerCustom prompt - project discovery and deduction agentno""
prompt_blueprintCustom prompt - blueprint agent (Chief Architect)no""
prompt_generatorCustom prompt - Dockerfile reasoning and creation agentno""
prompt_reviewerCustom prompt - security auditor and threat analysis agentno""
prompt_reflectorCustom prompt - detective and post-mortem analysis agentno""
prompt_error_analyzerCustom prompt - troubleshooter and error classification agentno""
prompt_iterative_improverCustom prompt - surgeon for precise corrections agentno""
llm_cachingEnable in-process LLM response caching (true/false)notrue
ollama_base_urlBase URL for Ollama when using llm_provider=ollamanohttp://localhost:11434
google_cloud_projectGoogle Cloud project ID for Gemini (optional)no""
collect_metricsEnable metrics collection (true/false)nofalse
enable_tracingEnable OpenTelemetry tracing (true/false)nofalse
tracing_exporterTracing exporter type (console, otlp)noconsole
otlp_endpointOTLP endpoint URL for trace exportnohttp://localhost:4317
otel_service_nameService name for OpenTelemetry tracesnodockai
langchain_tracing_v2Enable LangSmith/LangChain tracing (true/false)nofalse
langchain_api_keyLangSmith API key (required when langchain_tracing_v2=true)no
langchain_projectLangSmith project namenodockai
langchain_endpointLangSmith endpoint URLnohttps://api.smith.langchain.com

no outputs