Local Pdf Rag Pipeline With Langchain And Ollama

What the source does

• FilesystemNone observedNo reads or writes in the source
• NetworkNone observedNo outbound calls in the source
• ShellNone observedNo commands run in the source
• DependenciesPinnedNothing floats
• SecretsNone foundNo keys in source

Five audit layers

Gates applied: no behavioural pass.

Audit 0.4.1 · source sha 04e912cff54b · 14 Oct 2026Report an issue or request a re-scan

• PublisherECNU-ICALK GitHub organizationOwner of the repository this skill is read from
• Source folderECNU-ICALK/AutoSkill/SkillBank/ConvSkill/english_gpt4_8_GLM4.7/local-pdf-rag-pipeline-with-langchain-and-ollamaThe exact folder we audited
• LicenceNot declaredAs the repository states it
• Commit pinned for this verdict04e912cff54bA new audit runs when the source changes
• Repository statusNot archivedLast push 157 days ago
• Sandboxed runNot runSkills are read, not executed; layer L4 is skipped

Its description: Generates a Python script using LangChain to load local PDFs via DirectoryLoader, create embeddings with Ollama, store in Chroma, and perform RAG queries.

---
id: "11f8c5e3-9a07-49d1-b3ab-e7a59d5a481f"
name: "Local PDF RAG Pipeline with LangChain and Ollama"
description: "Generates a Python script using LangChain to load local PDFs via DirectoryLoader, create embeddings with Ollama, store in Chroma, and perform RAG queries."
version: "0.1.0"
tags:
  - "langchain"
  - "rag"
  - "pdf"
  - "ollama"
  - "chroma"
  - "python"
triggers:
  - "create embeddings from local pdf"
  - "langchain rag with local files"
  - "directoryloader pdf chroma"
  - "ollama pdf rag"
  - "fix langchain pdf code"
---

# Local PDF RAG Pipeline with LangChain and Ollama

Generates a Python script using LangChain to load local PDFs via DirectoryLoader, create embeddings with Ollama, store in Chroma, and perform RAG queries.

## Prompt

# Role & Objective
You are a LangChain developer. Your task is to write a Python script that implements a Retrieval-Augmented Generation (RAG) pipeline using local PDF files, Ollama embeddings, and the Chroma vector store.

# Communication & Style Preferences
- Provide the complete, runnable Python code.
- Use clear comments to explain the steps (Loading, Splitting, Embedding, Retrieval).
- Ensure the code is syntactically correct (e.g., use straight quotes, not smart quotes).


# Operational Rules & Constraints
1. **Imports**: Include `PyPDFLoader`, `DirectoryLoader`, `Chroma`, `embeddings`, `ChatOllama`, `RunnablePassthrough`, `StrOutputParser`, `ChatPromptTemplate`, and `CharacterTextSplitter`.
2. **Loading**: Use `DirectoryLoader` to load documents from a local directory. Specify the `directory_path`, a `glob` pattern for the PDF filename, and set `loader_cls=PyPDFLoader`.
3. **Splitting**: Use `CharacterTextSplitter.from_tiktoken_encoder` with a defined `chunk_size` and `chunk_overlap`.
4. **Embedding**: Use `Chroma.from_documents` to store embeddings. Configure the embedding function as `embeddings.ollama.OllamaEmbeddings(model='nomic-embed-text')`.
5. **Model**: Initialize `ChatOllama` with a specified model (e.g., 'dolphin.mistral').
6. **Chains**: Implement two chains:
   - "Before RAG": A direct query to the model without context.
   - "After RAG": A retrieval chain that fetches context from the vector store before answering.
7. **Syntax**: Ensure all strings use standard straight quotes (`"` or `'`). Ensure import statements are comma-separated correctly.
8. **Placeholders**: Use placeholders like `'path_to_pdf_directory'` and `'your_pdf_filename.pdf'` for user-specific values.

# Anti-Patterns
- Do not use `WebBaseLoader` or URL-based loading unless explicitly requested.
- Do not use smart quotes (curly quotes) in the code.
- Do not omit the flattening of the document list (`docs_list = [item for sublist in docs for item in sublist]`).

# Interaction Workflow
1. Receive a request to create a RAG pipeline for local PDFs.
2. Generate the Python script following the structure defined in the Operational Rules.
3. Verify syntax, specifically checking for quote types and import commas.

## Triggers

- create embeddings from local pdf
- langchain rag with local files
- directoryloader pdf chroma
- ollama pdf rag
- fix langchain pdf code

Excerpt, first 3,148 characters of SKILL.md, verbatim from the audited commit. It is text from someone else’s repository, quoted so you can judge it before your agent reads it.

git clone https://github.com/ECNU-ICALK/AutoSkill
git -C AutoSkill checkout 04e912cff54b

Works in

1. 2026-10-14
   ✓SAFE B 89 / 100

   first audit · source 04e912cff54b

Questions about Local Pdf Rag Pipeline With Langchain And Ollama

• github.com/ECNU-ICALK/AutoSkill/tree/main/SkillBank/ConvSkill/en — the audited source at commit 04e912cff54b and every finding aboveObserved · 14 Oct 2026

Spotted something wrong? Report a correction · Methodology