Add LangChain Deep Agents + MongoDB Atlas VFS tutorial - #230
Merged
mikikob-mongodb merged 2 commits intoSep 4, 2026
Merged
Conversation
Multi-agent research pipeline that searches a 45-file federal rulemaking docket (DOE air cleaner efficiency standards) across PDFs, spreadsheets, and DOCX files using hybrid $rankFusion search, then survives being killed mid-run via a manifest-based resume mechanism. Includes: - Corpus: DOE docket EERE-2021-BT-STD-0035 (public domain) - Scripts: seed, discovery, control arms (ripgrep + StoreBackend), pipeline runner with --kill-after, resume - Source: CompositeBackend wiring, agent definitions with production prompts, manifest system, metrics - Notebook: walkthrough.ipynb covering all three beats - Tests: deterministic golden answer acceptance test (regex, not LLM) - Tools: regulations.gov docket fetcher Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
GitHub blocks LFS uploads to public forks, preventing push. All binary files are well under the 100 MB per-file limit. Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
langchain-mongodb-deepagents-vfs)$rankFusionsearch across mixed formats (PDF, XLSX, DOCX), kill/resume durability via manifest.json run receipts, and a golden-answer acceptance testWhat's included
src/vfs_demo/— backend wiring, agent definitions, manifest system, metricsscripts/— seed corpus, control arms (ripgrep + StoreBackend), discovery, pipeline, resumenotebook/walkthrough.ipynb— end-to-end tutorial notebook (Beats 1-3)tests/golden_answer.py— deterministic regex acceptance test (6 checks)corpus/— vendored docket files (binary files via Git LFS)tools/fetch_docket.py— provenance script documenting how the corpus was assembledTest plan
python scripts/00_seed_corpus.pyseeds corpus to S3 and verifies searchabilitypython scripts/02_discovery.pyruns hybrid search with 3x reproducibilitypython scripts/03_pipeline.py --run-id test-001completes all 4 stagespython scripts/03_pipeline.py --run-id kill-001 --kill-after 2+python scripts/04_resume.py --run-id kill-001resumes from checkpointpython tests/golden_answer.py --run-id kill-001passes 6/6 golden-answer checkspapermill notebook/walkthrough.ipynb🤖 Generated with Claude Code