Replace llama-index with a local Document dataclass

Upgrade backend dependencies; adapt to langchain 1.3.15 and ruff 0.16

Archive private KB bundles for fast cold starts

Fix stale claims across repo docs

Improve KB command ergonomics, tool-output streaming, and wiki linting

docs: staleness sweep for core and pipeline docs

docs(evals): findings F35-F38, battery v2 confound corrections, staleness fixes

fix(retrieval): store the BM25 index as gzipped JSON instead of pickle (#13)

chore(versions): capture 2026-07-06 docs refresh, purge retired openai_cookbooks

fix(versions): llms.txt sources get version=null, not a Last-Modified date

fix(pipeline): single-source runs rebuild all_sources_data.jsonl safely

fix(pipeline): make upload_data_to_hf report failure honestly

fix(registry): retire sources from every registry dict, not just the first

fix(public-bundle): fail closed on non-allowlisted raw and wiki content

fix(pipeline): re-embed chunks whose content changed in place

fix(pipeline): stop incremental context runs gutting updated sources

Public docs-only bundle: allowlist filtering, leak audit, kb.tar.gz cold start (#11)

wip: eval harness + public docs bundle; relocate eval docs into evals/ (#10)

feat(registry): rename agentic_ai_engineering label to "Agent Engineering: Building Multi-Agent Systems"

feat(kb): register guardrails and serving topic pages in wiki scaffolder

Add 10-Hour LLM Primer course as a new KB source

Merge experiment #5: context-compaction study (keep-all vs compaction vs retrieve, Gemini 2.5 Flash)

Merge experiment #2: GraphRAG vs classical RAG (opt-in retriever + scoped comparison)

compaction study: harness for compaction-vs-keep-everything on a long lesson

graphrag: publish the index to HF for reproducibility (prod cold-start skips it)

graphrag: route Gemini 2.5 via OpenAI-compat endpoint; add index retry wrapper

graphrag: validate retriever vs real index, fix bugs, scope by source

Add OpenRouter provider for open-model evals

graphrag: scaffold true GraphRAG retriever arm (opt-in, default unchanged)

Refine v2 eval plan and memory harness

Add Part C eval variants, LLM judge, and harness fixes

Add evaluation harness for memory/context-management experiments

Enhance knowledge base and API functionality

Refactor project structure and update dependencies

Renamed 'scripts' folder to 'app', removed mongo and instructor libraries

Update project structure

ruff format

Refactor update_kb_wiki to remove --changed-only option

Remove .md extension from url from llms.txt sources

Revise kb tool prompt

shorter comments

add knowledge base wiki exploration tool

Update dependencies in pyproject.toml and uv.lock, including pytest and version upgrades for several packages. Modify .gitignore to exclude additional data directories and files. Enhance source_versions.json with new versioning information for various libraries.

Refactor scraping scripts to enhance source management and contextual node processing. Introduce new helper functions for pruning inactive contextual nodes and managing llms.txt sources. Update existing scripts to utilize the new source registry for improved source handling and streamline the process of adding and updating courses. Enhance error handling and logging for better traceability during data processing.

Remove "8 Hour Primer" references from scraping scripts and documentation, and introduce a new workflow for retiring sources from the data pipeline. Remove 8 Hour Primer from vector DB

Add Docker support with .dockerignore and Dockerfile updates. Integrate Next.js frontend build process, including package.json and related configurations. Enhance API to serve static frontend files and update README for project clarity. Add course and document metadata structures for improved content management.

Add source version capturing script and integrate it into the documentation update workflow. Enhance API to load source versions for frontend display, ensuring the latest release tags and commit SHAs are recorded for each source.

Update Full Stack and python courses

Add URL matching procedure documentation for course JSONL files, detailing the steps for scraping sidebar lessons, fuzzy matching, manual reconciliation, and final cleanup of unmatched rows.

Add Agentic AI Engineering course