Lightweight Implementation includes the DS-STAR Intelligence Layer with custom agents in about 1500 lines of code. Lightweight Implementation uses SQLAgent for text-to-SQL functionality. Lightweight Implementation uses RAG Agent for document question answering without heavy dependencies.

DataLens needs to adopt LiteLLM or LangChain for provider abstraction to support multiple LLM backends. DataLens requires using LiteLLM or similar abstraction to enable LLM flexibility.

IronClaw Agent Findings Visualization uses the Live Backend to process and visualize data. The Live Backend processes Qwen3 multi-block XML responses for SQL extraction. The Live Backend supports PostgreSQL Decimal type for findings generation.

Document RAG integrates with LlamaIndex. Document RAG uses LlamaIndex for document chunking and embedding generation. Document RAG capability integrates with LlamaIndex external system for document indexing and retrieval. Document RAG uses LlamaIndex with Qdrant for semantic document indexing. LlamaIndex generates embeddings using bge-large-en-v1.5 model. RAGAgent uses LlamaIndex to chunk PDFs and generate embeddings.

WrenAI supports multiple LLM providers including OpenAI, Anthropic, Ollama, and Bedrock.

The Multilingual Support (Danish) epic includes the LLM Prompt Injection use case. The LLM Prompt Injection use case modifies prompts in catalog.py to include language parameter for file summaries. The LLM Prompt Injection use case modifies prompts in question_router.py to include language parameter for answer synthesis.

Middleware components that intercept language model calls for caching, cost tracking, and prompt management.

No specific findings or updates provided in messages.

Pattern used for quality assessment and verification within the data analysis pipeline.

PHASE2_SCOPE_DECISION.md contains the LOAD_STRATEGY.md design element. Phase 2 research outputs include LOAD_STRATEGY.md as a component of research.

LocalAgentClient is used by IronClaw Agent Feature to maintain session state across API calls. LocalAgentClient uses SkillExecutor to process messages locally with an async generator. IronClaw Agent falls back to using LocalAgentClient when IronClaw environment variables are missing, resulting in local message processing and inability to complete queries successfully. theo backend uses LocalAgentClient if IRONCLAW_MODE environment variable is missing, leading to incomplete query processing in IronClaw Agent. LocalAgentClient.send_message uses SkillExecutor to execute and stream query results asynchronously. LocalAgentClient instantiates SkillExecutor internally for processing agent messages. LocalAgentClient.send_message uses SkillExecutor to execute and stream query results asynchronously. LocalAgentClient instantiates SkillExecutor internally for processing agent messages. LocalAgentClient directly uses SkillExecutor to run agent logic without IronClaw service dependency.

Logging is needed for findings generation to trace data flow and identify failure points during findings creation. Logging is necessary when AgentFinding is created to verify that findings are saved and streamed to the frontend properly. Logging is associated with SQL extraction regex fix as it will be added to trace extraction and execution results. question_router logging is a specific logging to be added after SQL query execution to provide diagnostics. findings_generator logging is a specific logging to be added at the start of findings generation for diagnostics. agent.py logging is a specific logging to be added when creating AgentFinding records to confirm saving of findings. Skill Logging records are stored in the agent_skill_log database table capturing skill execution details. Logging will be added to question_router to trace execution after SQL query to help diagnose findings generation issues. Logging will be added to findings_generator at the start of generate_findings to monitor numeric columns detection and diagnose failure points. Logging will be added in agent.py when AgentFinding records are created to verify findings saving to the database. The Logging framework instruments agent_skills.py process_message() to capture errors and execution flow.

The Frontend contains a Login page component. The Frontend UI Translation use case includes providing a language selector on the Login page. login page is implemented as a SvelteKit page. login page is part of the DataLens system. The login page is part of the DataLens project.

Logs are being used to identify silent failures in the async generator related to query execution, with recent updates adding detailed debug output to trace the exact error points.

npm dependency: lucide-svelte@^0.575.0, used in frontend, version 0.575.0, licensing and approval status unspecified.

Lux-api depends on pandas for auto-visualization features.

DataLens depends on the main agent for handling TerrainLens and for infrastructure coordination via @MoltBot. The main agent relies on @MoltBot as the infrastructure coordinator for requests and change approvals.

DataLens Development includes the achievement of the Major Milestone: Complete E2E Flow Working. The Complete E2E Flow Milestone utilizes several API Endpoints as part of the full user journey.

The Commit identification and deployment process involves using the Master branch for pushing the reverted commit. The last known good commit will be reverted, pushed to master branch and auto-deployed via Coolify as the next step to restore function.

A declarative schema layer used in other architectures for data modeling, not primary to DataLens.

PHASE2_SCOPE_DECISION.md includes the Memory decision point document dated 2026-02-25. PHASE2_SCOPE_DECISION.md includes the Memory decision point document as part of the Phase 2 research outputs. Phase 2 research outputs include the Memory decision point document as part of the documentation.

PHASE2_SCOPE_DECISION.md includes the Memory research document dated 2026-02-24.

Agent configuration includes the MEMORY.md file.