Integrations
parsr fits the stack you already run.
parsr is an HTTPS API first; everything else is a thin wrapper. Pick your agent framework, pick your vector database — they all speak the same `Chunk` shape and inherit parsr's region-binding and EU-residency story.
Frameworks
Agent runtimes
Drop parsr into your agent in three lines. Each package wraps the same five tools (parse_bank_statement, extract_with_schema, classify, get_job, get_usage), so the developer experience stays consistent across runtimes.
LangChain
Add parsr to any LangChain agent in three lines via `langchain-parsr`.
Read the guideLlamaIndex
Use parsr from any LlamaIndex agent via `llama-index-tool-parsr`.
Read the guideMCP (Model Context Protocol)
parsr as an MCP server — for Claude Desktop, Cursor, Continue, and any MCP-aware agent.
Read the guideVector databases
RAG over financial documents
Set include_chunks=true on any parse call to get RAG-friendly chunks alongside the structured payload — no extra billing. Each guide walks through parse → embed → upsert → query with the cost numbers you'll actually pay.
Pinecone
Index parsr-extracted financial documents in Pinecone for sub-100ms RAG retrieval.
Read the guideWeaviate
RAG over parsr-extracted documents with Weaviate's hybrid search and managed embeddings.
Read the guidepgvector (Postgres)
Add semantic search to your existing Postgres in 50 lines — parsr chunks + pgvector + HNSW.
Read the guideElasticsearch
Hybrid search over parsr-extracted finance docs with Elasticsearch's BM25 + dense vector RRF.
Read the guideQdrant
Self-hosted-or-managed vector DB with Rust performance — parsr chunks + Qdrant + payload filters.
Read the guideChroma
The fastest local-dev vector DB — Chroma + parsr chunks for prototypes and edge deployments.
Read the guideWorkflow tools
Make · Zapier · n8n
Direct parsr connectors are on the roadmap. Today the API is a plain HTTPS POST — every workflow tool can call it via their generic HTTP step. We won't build a workflow product (Make / Zapier / n8n already do that well); we'll publish reference templates as the integration audience grows.
Same API. Every integration. Pick your stack.
Browse the products