docs: add integration guide for PlainID SDK

[mottig30]

Description:
This PR adds a usage notebook for the PlainID integration with LangChain. It includes examples demonstrating how to build chains for data categorization, anonymization, and retrieval based on PlainID policy enforcement. This provides a practical guide for developers looking to embed authorization-aware agents using LangChain.

Issue:
None

Dependencies:
None

[vercel]

The latest updates on your projects. Learn more about Vercel for GitHub.

Project	Deployment	Preview	Comments	Updated (UTC)
langchain	Ready	Preview	Comment	Sep 12, 2025 8:51pm

[codspeed-hq]

CodSpeed WallTime Performance Report

Merging #32129 will improve performances by 19.28%

_{Comparing mottig30:patch-2 (3d59799) with master (338d3d2)}

⚠️ Unknown Walltime execution environment detected

Using the Walltime instrument on standard Hosted Runners will lead to inconsistent data.

For the most accurate results, we recommend using CodSpeed Macro Runners: bare-metal machines fine-tuned for performance measurement consistency.

Summary

⚡ 2 improvements
✅ 11 untouched

Benchmarks breakdown

	Benchmark	BASE	HEAD	Change
⚡	test_import_time[InMemoryVectorStore]	688.8 ms	594.6 ms	+15.86%
⚡	test_import_time[RunnableLambda]	560.7 ms	470 ms	+19.28%

[codspeed-hq]

CodSpeed Instrumentation Performance Report

Merging #32129 will not alter performance

_{Comparing mottig30:patch-2 (3d59799) with master (338d3d2)}

Summary

✅ 14 untouched

[Copilot]

Pull Request Overview

This PR adds comprehensive documentation for integrating PlainID SDK with LangChain, providing developers with practical guidance on implementing authorization-aware agents. The documentation covers the complete workflow from installation to building chains that handle data categorization, anonymization, and secure retrieval.

Key changes:

• Complete integration guide with PlainID setup instructions
• Examples for category filtering, data anonymization, and retrieval with policy enforcement
• Code examples demonstrating chain composition and vector store integration

Comments suppressed due to low confidence (3)

docs/docs/integrations/tools/plainid.ipynb:1

• The placeholder '' is invalid Python syntax and will cause a syntax error. This should be replaced with a proper classifier instance.

{"cells": [{"cell_type": "markdown", "id": "f77a88b0", "metadata": {}, "source": ["# LangChain PlainID Integration Guide"]}, {"cell_type": "markdown", "id": "d3a825b0", "metadata": {}, "source": ["## Installation"]}, {"cell_type": "markdown", "id": "91f1382a", "metadata": {}, "source": ["Based on your environment, you can install the library using pip:"]}, {"cell_type": "code", "execution_count": null, "id": "b3aefa28", "metadata": {}, "outputs": [], "source": ["!pip install langchain_plainid"]}, {"cell_type": "markdown", "id": "d05905bd", "metadata": {}, "source": ["## Setup with PlainID"]}, {"cell_type": "markdown", "id": "63146826", "metadata": {}, "source": ["1. Retrieve your PlainID credentials to access the platform - client ID and client secret.\n", "2. Find your PlainID base URL. For production use: `https://platform-product.us1.plainid.io` (note: starts with `platform-product.`)\n", "\n", "**Security Note:** Do not store credentials in code. Use environment variables or secret managers."]}, {"cell_type": "markdown", "id": "7fbc84dc", "metadata": {}, "source": ["## Category Filtering Setup in PlainID"]}, {"cell_type": "markdown", "id": "977d6d86", "metadata": {}, "source": ["You need to configure the relevant Ruleset in PlainID. For example, for template `categories`:\n", "```rego\n", "# METADATA\n", "# custom:\n", "#   plainid:\n", "#     kind: Ruleset\n", "#     name: All\n", "ruleset(asset, identity, requestParams, action) if {\n", "    asset.template == \"categories\"\n", "}\n", "```\n", "Also configure asset types like `contract`, `HR` in PlainID."]}, {"cell_type": "code", "execution_count": null, "id": "993b425f", "metadata": {}, "outputs": [], "source": ["from langchain_plainid import PlainIDCategorizer, PlainIDPermissionsProvider\n", "\n", "permissions_provider = PlainIDPermissionsProvider(\n", "    client_id=\"your_client_id\",\n", "    client_secret=\"your_client_secret\",\n", "    base_url=\"https://platform-product.us1.plainid.io\",\n", "    plainid_categories_resource_type=\"categories\"\n", ")\n", "\n", "plainid_categorizer = PlainIDCategorizer(\n", "    classifier_provider=<classifier>,\n", "    permissions_provider=permissions_provider\n", ")\n", "\n", "query = \"I'd like to know the weather forecast for today\"\n", "result = plainid_categorizer.invoke(query)"]}, {"cell_type": "markdown", "id": "f45ed4fd", "metadata": {}, "source": ["## Category Classifiers"]}, {"cell_type": "markdown", "id": "f9536725", "metadata": {}, "source": ["### 1. LLMCategoryClassifierProvider"]}, {"cell_type": "code", "execution_count": null, "id": "a3d8a816", "metadata": {}, "outputs": [], "source": ["from langchain_plainid import LLMCategoryClassifierProvider\n", "llm_classifier = LLMCategoryClassifierProvider(llm=OllamaLLM(model=\"llama2\"))"]}, {"cell_type": "markdown", "id": "c24e92b1", "metadata": {}, "source": ["### 2. ZeroShotCategoryClassifierProvider"]}, {"cell_type": "code", "execution_count": null, "id": "601563bd", "metadata": {}, "outputs": [], "source": ["from langchain_plainid import ZeroShotCategoryClassifierProvider\n", "zeroshot_classifier = ZeroShotCategoryClassifierProvider()"]}, {"cell_type": "markdown", "id": "ed0b6a48", "metadata": {}, "source": ["## Anonymizer Setup in PlainID"]}, {"cell_type": "markdown", "id": "4c215a0f", "metadata": {}, "source": ["Example Ruleset for `entities` template:\n", "```rego\n", "# METADATA\n", "# custom:\n", "#   plainid:\n", "#     kind: Ruleset\n", "#     name: PERSON\n", "ruleset(asset, identity, requestParams, action) if {\n", "    asset.template == \"entities\"\n", "    asset[\"path\"] == \"PERSON\"\n", "    action.id in [\"MASK\"]\n", "}\n", "```"]}, {"cell_type": "code", "execution_count": null, "id": "40de20d5", "metadata": {}, "outputs": [], "source": ["from langchain_plainid import PlainIDPermissionsProvider, PlainIDAnonymizer\n", "\n", "permissions_provider = PlainIDPermissionsProvider(\n", "    client_id=\"your_client_id\",\n", "    client_secret=\"your_client_secret\",\n", "    base_url=\"https://platform-product.us1.plainid.io\",\n", "    plainid_entities_resource_type=\"entities\"\n", ")\n", "\n", "plainid_anonymizer = PlainIDAnonymizer(\n", "    permissions_provider=permissions_provider,\n", "    encrypt_key=\"your_encryption_key\"\n", ")\n", "\n", "query = \"What's the name of the person who is responsible for the contract?\"\n", "result = plainid_anonymizer.invoke(query)"]}, {"cell_type": "markdown", "id": "cfa9b772", "metadata": {}, "source": ["## Full Chain Example"]}, {"cell_type": "code", "execution_count": null, "id": "7cb60e73", "metadata": {}, "outputs": [], "source": ["chain = plainid_categorizer | llm | vector_store | plainid_anonymizer | output_parser"]}, {"cell_type": "markdown", "id": "d43010fa", "metadata": {}, "source": ["## Retriever Setup"]}, {"cell_type": "markdown", "id": "ea6f43ce", "metadata": {}, "source": ["Rulesets for customer template:\n", "```rego\n", "ruleset(asset, identity, requestParams, action) if {\n", "    asset.template == \"customer\"\n", "    asset[\"country\"] == \"Sweden\"\n", "    asset[\"country\"] != \"Russia\"\n", "    contains(asset[\"country\"], \"we\")\n", "    startswith(asset[\"country\"], \"Sw\")\n", "}\n", "ruleset(asset, identity, requestParams, action) if {\n", "    asset.template == \"customer\"\n", "    asset[\"country\"] in [\"aaa\", \"bbb\"]\n", "    asset[\"age\"] <= 11111\n", "    endswith(asset[\"country\"], \"wwww\")\n", "}\n", "```"]}, {"cell_type": "code", "execution_count": null, "id": "8b3a41af", "metadata": {}, "outputs": [], "source": ["from langchain_community.vectorstores import Chroma\n", "from langchain_core.documents import Document\n", "from langchain_plainid import PlainIDRetriever\n", "\n", "documents = [\n", "    Document(\"Stockholm is the capital of Sweden.\", metadata={\"country\": \"Sweden\", \"age\": 5}),\n", "    Document(\"Oslo is the capital of Norway.\", metadata={\"country\": \"Norway\", \"age\": 5}),\n", "    Document(\"Copenhagen is the capital of Denmark.\", metadata={\"country\": \"Denmark\", \"age\": 5}),\n", "    Document(\"Helsinki is the capital of Finland.\", metadata={\"country\": \"Finland\", \"age\": 5}),\n", "    Document(\"Malm\u00f6 is a city in Sweden.\", metadata={\"country\": \"Sweden\", \"age\": 5}),\n", "]\n", "\n", "vector_store = Chroma.from_documents(documents, embeddings)\n", "plainid_retriever = PlainIDRetriever(vectorstore=vector_store, filter_provider=filter_provider)\n", "docs = plainid_retriever.invoke(\"What is the capital of Sweden?\")"]}, {"cell_type": "markdown", "id": "b0b3ceb8", "metadata": {}, "source": ["## Supported Vector Stores and Limitations"]}, {"cell_type": "markdown", "id": "11f1ec03", "metadata": {}, "source": ["| Vector Store | Not Supported Operators                       |\n", "|--------------|-----------------------------------------------|\n", "| FAISS        | STARTSWITH, ENDSWITH, CONTAINS                |\n", "| Chroma       | IN, NOT_IN, STARTSWITH, ENDSWITH, CONTAINS    |"]}], "metadata": {"title": "LangChain PlainID Integration"}, "nbformat": 4, "nbformat_minor": 5}

docs/docs/integrations/tools/plainid.ipynb:1

• OllamaLLM is used without importing it. This will result in a NameError at runtime.

{"cells": [{"cell_type": "markdown", "id": "f77a88b0", "metadata": {}, "source": ["# LangChain PlainID Integration Guide"]}, {"cell_type": "markdown", "id": "d3a825b0", "metadata": {}, "source": ["## Installation"]}, {"cell_type": "markdown", "id": "91f1382a", "metadata": {}, "source": ["Based on your environment, you can install the library using pip:"]}, {"cell_type": "code", "execution_count": null, "id": "b3aefa28", "metadata": {}, "outputs": [], "source": ["!pip install langchain_plainid"]}, {"cell_type": "markdown", "id": "d05905bd", "metadata": {}, "source": ["## Setup with PlainID"]}, {"cell_type": "markdown", "id": "63146826", "metadata": {}, "source": ["1. Retrieve your PlainID credentials to access the platform - client ID and client secret.\n", "2. Find your PlainID base URL. For production use: `https://platform-product.us1.plainid.io` (note: starts with `platform-product.`)\n", "\n", "**Security Note:** Do not store credentials in code. Use environment variables or secret managers."]}, {"cell_type": "markdown", "id": "7fbc84dc", "metadata": {}, "source": ["## Category Filtering Setup in PlainID"]}, {"cell_type": "markdown", "id": "977d6d86", "metadata": {}, "source": ["You need to configure the relevant Ruleset in PlainID. For example, for template `categories`:\n", "```rego\n", "# METADATA\n", "# custom:\n", "#   plainid:\n", "#     kind: Ruleset\n", "#     name: All\n", "ruleset(asset, identity, requestParams, action) if {\n", "    asset.template == \"categories\"\n", "}\n", "```\n", "Also configure asset types like `contract`, `HR` in PlainID."]}, {"cell_type": "code", "execution_count": null, "id": "993b425f", "metadata": {}, "outputs": [], "source": ["from langchain_plainid import PlainIDCategorizer, PlainIDPermissionsProvider\n", "\n", "permissions_provider = PlainIDPermissionsProvider(\n", "    client_id=\"your_client_id\",\n", "    client_secret=\"your_client_secret\",\n", "    base_url=\"https://platform-product.us1.plainid.io\",\n", "    plainid_categories_resource_type=\"categories\"\n", ")\n", "\n", "plainid_categorizer = PlainIDCategorizer(\n", "    classifier_provider=<classifier>,\n", "    permissions_provider=permissions_provider\n", ")\n", "\n", "query = \"I'd like to know the weather forecast for today\"\n", "result = plainid_categorizer.invoke(query)"]}, {"cell_type": "markdown", "id": "f45ed4fd", "metadata": {}, "source": ["## Category Classifiers"]}, {"cell_type": "markdown", "id": "f9536725", "metadata": {}, "source": ["### 1. LLMCategoryClassifierProvider"]}, {"cell_type": "code", "execution_count": null, "id": "a3d8a816", "metadata": {}, "outputs": [], "source": ["from langchain_plainid import LLMCategoryClassifierProvider\n", "llm_classifier = LLMCategoryClassifierProvider(llm=OllamaLLM(model=\"llama2\"))"]}, {"cell_type": "markdown", "id": "c24e92b1", "metadata": {}, "source": ["### 2. ZeroShotCategoryClassifierProvider"]}, {"cell_type": "code", "execution_count": null, "id": "601563bd", "metadata": {}, "outputs": [], "source": ["from langchain_plainid import ZeroShotCategoryClassifierProvider\n", "zeroshot_classifier = ZeroShotCategoryClassifierProvider()"]}, {"cell_type": "markdown", "id": "ed0b6a48", "metadata": {}, "source": ["## Anonymizer Setup in PlainID"]}, {"cell_type": "markdown", "id": "4c215a0f", "metadata": {}, "source": ["Example Ruleset for `entities` template:\n", "```rego\n", "# METADATA\n", "# custom:\n", "#   plainid:\n", "#     kind: Ruleset\n", "#     name: PERSON\n", "ruleset(asset, identity, requestParams, action) if {\n", "    asset.template == \"entities\"\n", "    asset[\"path\"] == \"PERSON\"\n", "    action.id in [\"MASK\"]\n", "}\n", "```"]}, {"cell_type": "code", "execution_count": null, "id": "40de20d5", "metadata": {}, "outputs": [], "source": ["from langchain_plainid import PlainIDPermissionsProvider, PlainIDAnonymizer\n", "\n", "permissions_provider = PlainIDPermissionsProvider(\n", "    client_id=\"your_client_id\",\n", "    client_secret=\"your_client_secret\",\n", "    base_url=\"https://platform-product.us1.plainid.io\",\n", "    plainid_entities_resource_type=\"entities\"\n", ")\n", "\n", "plainid_anonymizer = PlainIDAnonymizer(\n", "    permissions_provider=permissions_provider,\n", "    encrypt_key=\"your_encryption_key\"\n", ")\n", "\n", "query = \"What's the name of the person who is responsible for the contract?\"\n", "result = plainid_anonymizer.invoke(query)"]}, {"cell_type": "markdown", "id": "cfa9b772", "metadata": {}, "source": ["## Full Chain Example"]}, {"cell_type": "code", "execution_count": null, "id": "7cb60e73", "metadata": {}, "outputs": [], "source": ["chain = plainid_categorizer | llm | vector_store | plainid_anonymizer | output_parser"]}, {"cell_type": "markdown", "id": "d43010fa", "metadata": {}, "source": ["## Retriever Setup"]}, {"cell_type": "markdown", "id": "ea6f43ce", "metadata": {}, "source": ["Rulesets for customer template:\n", "```rego\n", "ruleset(asset, identity, requestParams, action) if {\n", "    asset.template == \"customer\"\n", "    asset[\"country\"] == \"Sweden\"\n", "    asset[\"country\"] != \"Russia\"\n", "    contains(asset[\"country\"], \"we\")\n", "    startswith(asset[\"country\"], \"Sw\")\n", "}\n", "ruleset(asset, identity, requestParams, action) if {\n", "    asset.template == \"customer\"\n", "    asset[\"country\"] in [\"aaa\", \"bbb\"]\n", "    asset[\"age\"] <= 11111\n", "    endswith(asset[\"country\"], \"wwww\")\n", "}\n", "```"]}, {"cell_type": "code", "execution_count": null, "id": "8b3a41af", "metadata": {}, "outputs": [], "source": ["from langchain_community.vectorstores import Chroma\n", "from langchain_core.documents import Document\n", "from langchain_plainid import PlainIDRetriever\n", "\n", "documents = [\n", "    Document(\"Stockholm is the capital of Sweden.\", metadata={\"country\": \"Sweden\", \"age\": 5}),\n", "    Document(\"Oslo is the capital of Norway.\", metadata={\"country\": \"Norway\", \"age\": 5}),\n", "    Document(\"Copenhagen is the capital of Denmark.\", metadata={\"country\": \"Denmark\", \"age\": 5}),\n", "    Document(\"Helsinki is the capital of Finland.\", metadata={\"country\": \"Finland\", \"age\": 5}),\n", "    Document(\"Malm\u00f6 is a city in Sweden.\", metadata={\"country\": \"Sweden\", \"age\": 5}),\n", "]\n", "\n", "vector_store = Chroma.from_documents(documents, embeddings)\n", "plainid_retriever = PlainIDRetriever(vectorstore=vector_store, filter_provider=filter_provider)\n", "docs = plainid_retriever.invoke(\"What is the capital of Sweden?\")"]}, {"cell_type": "markdown", "id": "b0b3ceb8", "metadata": {}, "source": ["## Supported Vector Stores and Limitations"]}, {"cell_type": "markdown", "id": "11f1ec03", "metadata": {}, "source": ["| Vector Store | Not Supported Operators                       |\n", "|--------------|-----------------------------------------------|\n", "| FAISS        | STARTSWITH, ENDSWITH, CONTAINS                |\n", "| Chroma       | IN, NOT_IN, STARTSWITH, ENDSWITH, CONTAINS    |"]}], "metadata": {"title": "LangChain PlainID Integration"}, "nbformat": 4, "nbformat_minor": 5}

docs/docs/integrations/tools/plainid.ipynb:1

• [nitpick] While the security note mentions not storing credentials in code, the examples still use placeholder strings that could be mistakenly used as-is. Consider using environment variable examples like os.getenv('PLAINID_CLIENT_ID') to demonstrate best practices.

{"cells": [{"cell_type": "markdown", "id": "f77a88b0", "metadata": {}, "source": ["# LangChain PlainID Integration Guide"]}, {"cell_type": "markdown", "id": "d3a825b0", "metadata": {}, "source": ["## Installation"]}, {"cell_type": "markdown", "id": "91f1382a", "metadata": {}, "source": ["Based on your environment, you can install the library using pip:"]}, {"cell_type": "code", "execution_count": null, "id": "b3aefa28", "metadata": {}, "outputs": [], "source": ["!pip install langchain_plainid"]}, {"cell_type": "markdown", "id": "d05905bd", "metadata": {}, "source": ["## Setup with PlainID"]}, {"cell_type": "markdown", "id": "63146826", "metadata": {}, "source": ["1. Retrieve your PlainID credentials to access the platform - client ID and client secret.\n", "2. Find your PlainID base URL. For production use: `https://platform-product.us1.plainid.io` (note: starts with `platform-product.`)\n", "\n", "**Security Note:** Do not store credentials in code. Use environment variables or secret managers."]}, {"cell_type": "markdown", "id": "7fbc84dc", "metadata": {}, "source": ["## Category Filtering Setup in PlainID"]}, {"cell_type": "markdown", "id": "977d6d86", "metadata": {}, "source": ["You need to configure the relevant Ruleset in PlainID. For example, for template `categories`:\n", "```rego\n", "# METADATA\n", "# custom:\n", "#   plainid:\n", "#     kind: Ruleset\n", "#     name: All\n", "ruleset(asset, identity, requestParams, action) if {\n", "    asset.template == \"categories\"\n", "}\n", "```\n", "Also configure asset types like `contract`, `HR` in PlainID."]}, {"cell_type": "code", "execution_count": null, "id": "993b425f", "metadata": {}, "outputs": [], "source": ["from langchain_plainid import PlainIDCategorizer, PlainIDPermissionsProvider\n", "\n", "permissions_provider = PlainIDPermissionsProvider(\n", "    client_id=\"your_client_id\",\n", "    client_secret=\"your_client_secret\",\n", "    base_url=\"https://platform-product.us1.plainid.io\",\n", "    plainid_categories_resource_type=\"categories\"\n", ")\n", "\n", "plainid_categorizer = PlainIDCategorizer(\n", "    classifier_provider=<classifier>,\n", "    permissions_provider=permissions_provider\n", ")\n", "\n", "query = \"I'd like to know the weather forecast for today\"\n", "result = plainid_categorizer.invoke(query)"]}, {"cell_type": "markdown", "id": "f45ed4fd", "metadata": {}, "source": ["## Category Classifiers"]}, {"cell_type": "markdown", "id": "f9536725", "metadata": {}, "source": ["### 1. LLMCategoryClassifierProvider"]}, {"cell_type": "code", "execution_count": null, "id": "a3d8a816", "metadata": {}, "outputs": [], "source": ["from langchain_plainid import LLMCategoryClassifierProvider\n", "llm_classifier = LLMCategoryClassifierProvider(llm=OllamaLLM(model=\"llama2\"))"]}, {"cell_type": "markdown", "id": "c24e92b1", "metadata": {}, "source": ["### 2. ZeroShotCategoryClassifierProvider"]}, {"cell_type": "code", "execution_count": null, "id": "601563bd", "metadata": {}, "outputs": [], "source": ["from langchain_plainid import ZeroShotCategoryClassifierProvider\n", "zeroshot_classifier = ZeroShotCategoryClassifierProvider()"]}, {"cell_type": "markdown", "id": "ed0b6a48", "metadata": {}, "source": ["## Anonymizer Setup in PlainID"]}, {"cell_type": "markdown", "id": "4c215a0f", "metadata": {}, "source": ["Example Ruleset for `entities` template:\n", "```rego\n", "# METADATA\n", "# custom:\n", "#   plainid:\n", "#     kind: Ruleset\n", "#     name: PERSON\n", "ruleset(asset, identity, requestParams, action) if {\n", "    asset.template == \"entities\"\n", "    asset[\"path\"] == \"PERSON\"\n", "    action.id in [\"MASK\"]\n", "}\n", "```"]}, {"cell_type": "code", "execution_count": null, "id": "40de20d5", "metadata": {}, "outputs": [], "source": ["from langchain_plainid import PlainIDPermissionsProvider, PlainIDAnonymizer\n", "\n", "permissions_provider = PlainIDPermissionsProvider(\n", "    client_id=\"your_client_id\",\n", "    client_secret=\"your_client_secret\",\n", "    base_url=\"https://platform-product.us1.plainid.io\",\n", "    plainid_entities_resource_type=\"entities\"\n", ")\n", "\n", "plainid_anonymizer = PlainIDAnonymizer(\n", "    permissions_provider=permissions_provider,\n", "    encrypt_key=\"your_encryption_key\"\n", ")\n", "\n", "query = \"What's the name of the person who is responsible for the contract?\"\n", "result = plainid_anonymizer.invoke(query)"]}, {"cell_type": "markdown", "id": "cfa9b772", "metadata": {}, "source": ["## Full Chain Example"]}, {"cell_type": "code", "execution_count": null, "id": "7cb60e73", "metadata": {}, "outputs": [], "source": ["chain = plainid_categorizer | llm | vector_store | plainid_anonymizer | output_parser"]}, {"cell_type": "markdown", "id": "d43010fa", "metadata": {}, "source": ["## Retriever Setup"]}, {"cell_type": "markdown", "id": "ea6f43ce", "metadata": {}, "source": ["Rulesets for customer template:\n", "```rego\n", "ruleset(asset, identity, requestParams, action) if {\n", "    asset.template == \"customer\"\n", "    asset[\"country\"] == \"Sweden\"\n", "    asset[\"country\"] != \"Russia\"\n", "    contains(asset[\"country\"], \"we\")\n", "    startswith(asset[\"country\"], \"Sw\")\n", "}\n", "ruleset(asset, identity, requestParams, action) if {\n", "    asset.template == \"customer\"\n", "    asset[\"country\"] in [\"aaa\", \"bbb\"]\n", "    asset[\"age\"] <= 11111\n", "    endswith(asset[\"country\"], \"wwww\")\n", "}\n", "```"]}, {"cell_type": "code", "execution_count": null, "id": "8b3a41af", "metadata": {}, "outputs": [], "source": ["from langchain_community.vectorstores import Chroma\n", "from langchain_core.documents import Document\n", "from langchain_plainid import PlainIDRetriever\n", "\n", "documents = [\n", "    Document(\"Stockholm is the capital of Sweden.\", metadata={\"country\": \"Sweden\", \"age\": 5}),\n", "    Document(\"Oslo is the capital of Norway.\", metadata={\"country\": \"Norway\", \"age\": 5}),\n", "    Document(\"Copenhagen is the capital of Denmark.\", metadata={\"country\": \"Denmark\", \"age\": 5}),\n", "    Document(\"Helsinki is the capital of Finland.\", metadata={\"country\": \"Finland\", \"age\": 5}),\n", "    Document(\"Malm\u00f6 is a city in Sweden.\", metadata={\"country\": \"Sweden\", \"age\": 5}),\n", "]\n", "\n", "vector_store = Chroma.from_documents(documents, embeddings)\n", "plainid_retriever = PlainIDRetriever(vectorstore=vector_store, filter_provider=filter_provider)\n", "docs = plainid_retriever.invoke(\"What is the capital of Sweden?\")"]}, {"cell_type": "markdown", "id": "b0b3ceb8", "metadata": {}, "source": ["## Supported Vector Stores and Limitations"]}, {"cell_type": "markdown", "id": "11f1ec03", "metadata": {}, "source": ["| Vector Store | Not Supported Operators                       |\n", "|--------------|-----------------------------------------------|\n", "| FAISS        | STARTSWITH, ENDSWITH, CONTAINS                |\n", "| Chroma       | IN, NOT_IN, STARTSWITH, ENDSWITH, CONTAINS    |"]}], "metadata": {"title": "LangChain PlainID Integration"}, "nbformat": 4, "nbformat_minor": 5}

[Copilot]

Multiple undefined variables are used in this chain: 'llm', 'vector_store', and 'output_parser'. These need to be defined or imported before use.

[Copilot]

The variable 'filter_provider' is used without being defined. This should be defined or replaced with the correct permissions_provider instance.

[Copilot]

The variable 'embeddings' is used without being defined or imported. An embeddings instance needs to be created first.

[mdrxy]

The notebook is malformed.

Sorry for the delay - we've been in the middle of a docs hiatus while we port things over to the new docs site.

These docs will soon become invisible as we deprecate the current docs pages.

Please view the new contributing guide for further info on how to port. Don't hesitate to ask questions, as we've rewritten parts of it and may have missed something.