DocPull MCP server

DocPull

Active open-source project · MIT License

DocPull turns changing public web sources into cited, reproducible context for AI agents and retrieval pipelines. Use it when your application needs to know which sources it used, whether they changed, and how to rebuild the same context later.

Install and sync your first source

pip install docpull
docpull init stripe-docs
docpull add https://docs.stripe.com
docpull sync
docpull query "retry policy" --max-tokens 4000
docpull diff
docpull export context-pack --target cursor

The project stores declared sources in docpull.yaml and resolved inputs in .docpull/context.lock.json. Later syncs produce a hash-based diff while preserving source URLs, content hashes, run IDs, citations, and export metadata.

docpull query retrieves deduplicated passages from locked archives with version filters, a response budget, and exact source spans. The same query is available through the Python SDK and MCP. See Query locked context for budget semantics and queries across projects.

No account or paid API is required for this path. Direct fetching, discovery, extraction, indexing, pack analysis, and diffs run locally.

Why use DocPull

• Reproduce agent context. Stable IDs, hashes, manifests, and lockfiles show which source versions produced an answer or artifact.
• Detect source drift. Sync and diff documentation, product pages, policies, feeds, repositories, packages, standards, and local documents.
• Keep evidence inspectable. Markdown, NDJSON, SQLite, citations, and provenance sidecars remain readable without a hosted service.
• Choose the downstream surface. Export context for agent clients, vector import, data workflows, or a versioned context-pack release.
• Keep expensive routes explicit. Browser and cloud rendering require an explicit choice and can be blocked with a zero-dollar budget.

How it works

declared sources → local acquisition → versioned evidence → diff and validation → export

DocPull's v3 pack contract separates raw extraction, agent-ready context, and eval-grade evidence. Validate the level a downstream system requires:

docpull pack prepare packs/docs --eval-grade
docpull pack validate packs/docs --level eval
docpull ci --prepare

docpull ci checks freshness, citation coverage, pack quality, rights metadata, and other configured gates. It writes context-ci.report.json and CONTEXT_CI.md, then exits non-zero when a hard gate fails.

Supported surfaces

Surface	Use it for	Start here
CLI	Fetch, sync, diff, validate, and export	docs/cli-recipes.md
Python SDK	Embed acquisition in Python applications	docs/surface-contract.md
MCP server	Give local agent clients source tools	MCP server
TypeScript SDK	Read local packs and invoke the CLI from Node or Bun	sdk/js/README.md
Agent plugin	Install the supported MCP workflow in an agent client	plugin/README.md

Common source shapes include static and server-rendered websites, OpenAPI documents, feeds, papers, public GitHub repositories, npm and PyPI packages, standards, datasets, transcripts, Wikimedia pages, product and policy pages, and local PDF or office files. See context-pack workflows for the complete surface.

MCP server

pip install 'docpull[mcp]'
docpull mcp

Claude Code can register the same local server:

claude mcp add --transport stdio docpull -- docpull mcp

The Python stdio server is the supported MCP server.

Limits and security boundary

DocPull is an evidence-acquisition engine, not a hosted competitive-intelligence product. It owns fetching, explicit rendering adapters, versioning, citations, hashing, validation, replay, and export. Downstream products own scheduling, human review, approved claims, legal conclusions, accounts, and notifications.

The default path does not handle complex interactive browser workflows, CAPTCHAs, stealth scraping, or private dashboards. JavaScript rendering is explicit. Authenticated sources require environment-variable references; DocPull does not persist credential values in project artifacts.

Security defaults include HTTPS-only fetching, robots.txt compliance, SSRF and DNS rebinding protections, redirect guards, XXE protection, and path-traversal checks. Read the web-source boundary, security posture, and evidence-engine decision before extending acquisition behavior.

Documentation and evidence

• CLI recipes — Common commands and advanced workflows.
• Context dependencies — Project and lockfile model.
• Context Pack Contract v3 — Artifact levels and compatibility.
• Public contracts — Schemas and versioning rules.
• Alternatives — Browser automation and hosted extraction tradeoffs.
• JavaScript SDK and loaders — TypeScript client and LangChain/LlamaIndex loaders.
• Dependency security — Locked dependency sets and optional parser support.
• Evaluation lab — Reproducible methods, results, and claim boundaries.
• Changelog — Release history.

Raintree open-source system

DocPull owns evidence acquisition and reproducible agent context. It can be used independently; the sibling projects do not imply a required integration or shared release cycle.

Project	Responsibility
Raintree Standards	Defines governed requirements and evidence.
Trellis	Enforces shared JavaScript and TypeScript code policy.
HIG Doctor	Audits interface source and provides HIG guidance.
PolicyStrata	Tests cross-layer policy behavior.

See the Raintree open-source portfolio for current lifecycle and distribution links.

Raintree also builds BetterOff, which keeps a household's spending, investments, assets, debts, and net worth in one current record.

Project policies

Contributing · Code of Conduct · Security · Metrics and evidence limits · Source repository · MIT License