Skip to content

The agent engineering platform.

11.0 KB
Raw Blame
{
  "schemaVersion": 1,
  "pageVersion": "sha256:fca157e721bb0e9c21d172bb3090cbacd7b1aece35a9708b749eacdfc2b6a3ec",
  "claims": [
    {
      "id": "claim_c71edae5b0f64c8fa79c17ef1aca0f6b",
      "statement": "The Agent Factory is implemented in factory.py, which constructs compiled LangGraph state machines for agentic loops that orchestrate model invocation, tool execution, and middleware composition.",
      "evidence": [
        {
          "resource": "repo://libs/langchain_v1/langchain/agents/factory.py",
          "version": "repo-file-v1:sha256:cc4a7d9a620224558c4eddd4e13edc988a9ac9e328b7a67deeb71ed2c5ec5174"
        }
      ]
    },
    {
      "id": "claim_0587839e54be45c8924d9154860a0345",
      "statement": "Agent middleware system is implemented in the middleware directory, providing pluggable hooks for injecting logic at model, tool, and lifecycle boundaries.",
      "evidence": [
        {
          "resource": "repo://libs/langchain_v1/langchain/agents/middleware/types.py",
          "version": "repo-file-v1:sha256:56688206dce0334e7ca82125d7b809c23427890c234dfd69123dad038a387f31"
        }
      ]
    },
    {
      "id": "claim_3ca5cc3aa2d44f1aaa11eafd8fa56fb9",
      "statement": "BaseChatModel abstract base class and unified provider interface are implemented in langchain_core/language_models/chat_models.py, handling streaming, batching, token counting, and structured output marshaling.",
      "evidence": [
        {
          "resource": "repo://libs/core/langchain_core/language_models/chat_models.py",
          "version": "repo-file-v1:sha256:7377a3e0d534d6fe9835cb35e39fc2bf7614ae2be26eb5664a0767dfb002cb03"
        }
      ]
    },
    {
      "id": "claim_f5a9f756fd1b4f3b843ef1551cfd2263",
      "statement": "Message types (AIMessage, HumanMessage, SystemMessage, ToolMessage) and content blocks are defined in langchain_core/messages/, providing unified message representation across all provider integrations.",
      "evidence": [
        {
          "resource": "repo://libs/core/langchain_core/messages/base.py",
          "version": "repo-file-v1:sha256:7398bb4db990af959df71dd81243e04b1a5f665602155a6f808271d910c4a62d"
        }
      ]
    },
    {
      "id": "claim_41da34bd04da40ffb6c3ac4c336123be",
      "statement": "The Runnable protocol is the foundational abstraction for composable units, implemented in langchain_core/runnables/base.py, defining invoke(), stream(), batch(), and async variants, with composition operators for piping, branching, fallback, and retry.",
      "evidence": [
        {
          "resource": "repo://libs/core/langchain_core/runnables/base.py",
          "version": "repo-file-v1:sha256:4cdcf9f699860ef49c02fe75339cdd8f302f050fa108b625f9a18d84cf3367ba"
        }
      ]
    },
    {
      "id": "claim_de27d6d3996d46669d55a5294a6cb6a7",
      "statement": "BaseTool abstraction and tool conversion utilities are implemented in langchain_core/tools/, supporting sync/async invocation, schema generation from docstrings and Pydantic models, and structured argument parsing.",
      "evidence": [
        {
          "resource": "repo://libs/core/langchain_core/tools/base.py",
          "version": "repo-file-v1:sha256:a2f7b4f18aaa3f6db1399f224f6303691cb46ac3b9ba36bf1aa87704a504edd1"
        }
      ]
    },
    {
      "id": "claim_b6a938d67cf04329abc2cce336bd6a08",
      "statement": "Prompt templates are implemented in langchain_core/prompts/, including chat prompts, few-shot examples, image prompts, and structured prompt abstractions.",
      "evidence": [
        {
          "resource": "repo://libs/core/langchain_core/prompts/base.py",
          "version": "repo-file-v1:sha256:f2f608fba240b0d596889a9ca96ca1f8f4f49690bc24abd094f1b87c776c7f30"
        }
      ]
    },
    {
      "id": "claim_4a3cb18e1ffc46afbc05b276bc1dda11",
      "statement": "Callback manager infrastructure for instrumentation and LangSmith integration is implemented in langchain_core/callbacks/, providing base handlers and streaming output callbacks.",
      "evidence": [
        {
          "resource": "repo://libs/core/langchain_core/callbacks/manager.py",
          "version": "repo-file-v1:sha256:323e48901ebf1ecd2c1b197e8cc1d81298366b29ee89aa02ab729d725fbcfebc"
        }
      ]
    },
    {
      "id": "claim_bb74dbecd28347a68d235dc352fd1fef",
      "statement": "The init_chat_model() factory function for dynamic model discovery and instantiation by provider:model identifier is implemented in langchain/chat_models/base.py.",
      "evidence": [
        {
          "resource": "repo://libs/langchain_v1/langchain/chat_models/base.py",
          "version": "repo-file-v1:sha256:757316423a5b1f95b14b1ebaac56e612af026bb50f06c1748e688e65a8a17b67"
        }
      ]
    },
    {
      "id": "claim_27f57c927a0e4ae3964bc8a835fbec5e",
      "statement": "Provider-specific implementations such as ChatOpenAI are located in libs/partners/<provider>/ directories, each implementing core abstractions and released independently.",
      "evidence": [
        {
          "resource": "repo://libs/partners/openai/langchain_openai/chat_models/base.py",
          "version": "repo-file-v1:sha256:50aec54e3bdec7e049a579eb53a1cdb999011889714a4f8b8b6be409b1cd025a"
        }
      ]
    },
    {
      "id": "claim_3b029c6ea6b141ea86e8c4f56420a82b",
      "statement": "LLM behavior and capability profiles are implemented in libs/model-profiles/langchain_model_profiles/, providing metadata about model capabilities and configuration.",
      "evidence": [
        {
          "resource": "repo://libs/model-profiles/langchain_model_profiles/__init__.py",
          "version": "repo-file-v1:sha256:e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855"
        }
      ]
    },
    {
      "id": "claim_00c705d2ed384205a4c87452106b2842",
      "statement": "Model Context Protocol (MCP) adapter, tools, and elicitation are implemented in langchain/mcp/, providing MCP-based model integrations.",
      "evidence": [
        {
          "resource": "repo://libs/langchain_v1/langchain/mcp/adapter.py",
          "version": "repo-file-v1:sha256:9a93363b3b18055d06778bf0508f63b85f382a829f9588b98257e014be45c88e"
        }
      ]
    },
    {
      "id": "claim_4debd15d45c545a798d12cf25ccfa03a",
      "statement": "Schema definition and response marshaling for typed agent outputs is implemented in langchain/agents/structured_output.py.",
      "evidence": [
        {
          "resource": "repo://libs/langchain_v1/langchain/agents/structured_output.py",
          "version": "repo-file-v1:sha256:e3d5032f156d07531da71620710f0bb0231b3ba8ab241a5417759ebb810b71cf"
        }
      ]
    },
    {
      "id": "claim_a663419859ee445c871c8b614e26dc85",
      "statement": "Tests are organized as unit_tests/ and integration_tests/ within each package (core, langchain_v1, partners), with shared test contracts in libs/standard-tests/.",
      "evidence": [
        {
          "resource": "repo://libs/core/tests/unit_tests/test_tools.py",
          "version": "repo-file-v1:sha256:18a0d8474908b28ccd2940f35efbb04e43dfd9d4df93c0a1ad2f65777a165ece"
        },
        {
          "resource": "repo://libs/langchain_v1/tests/unit_tests/test_imports.py",
          "version": "repo-file-v1:sha256:1ac3e45f3bfc2b66ab87e2b6d5902cc90570f0f207ce067c40a7943a6fadebe4"
        },
        {
          "resource": "repo://libs/standard-tests/langchain_tests/base.py",
          "version": "repo-file-v1:sha256:62a5d5e75b59be001165917da5164c60bbdb07394b364df2ea1b949916ba3cab"
        }
      ]
    },
    {
      "id": "claim_691d6cb116a343ed92077ddb6a6078bd",
      "statement": "Configuration files include pyproject.toml in each package for dependencies and metadata, pre-commit configuration at repo root for git hooks, and Makefile in libs/ for multi-package build coordination.",
      "evidence": [
        {
          "resource": "repo://.pre-commit-config.yaml",
          "version": "repo-file-v1:sha256:f6173b7f48573811a6a8ca9f0691ecb2d79c331ac1676af11542ab5cdcf68deb"
        },
        {
          "resource": "repo://libs/Makefile",
          "version": "repo-file-v1:sha256:b703a85cdfdb60e98c41ec8261d9b78f0e7137105ee9da36b8cb919b2d35858b"
        },
        {
          "resource": "repo://libs/core/pyproject.toml",
          "version": "repo-file-v1:sha256:8bdfc645213004979b0e0c0c60d53038efc82f72d29c9c316ecc8ad02f99f3fd"
        },
        {
          "resource": "repo://libs/langchain_v1/pyproject.toml",
          "version": "repo-file-v1:sha256:b996371876a59970935a484867bfcb2d5af624f5b1107ffe83f524c237597098"
        }
      ]
    },
    {
      "id": "claim_458d709b604149fa8fb9639025882358",
      "statement": "LangChain uses a three-layer architecture: langchain-core (v1.6.13) for base abstractions, langchain (v1.4.7) for orchestration and agents, and partners for provider-specific implementations, enabling model interoperability and independent release cycles.",
      "evidence": [
        {
          "resource": "repo://libs/core/pyproject.toml#L24-L24",
          "version": "repo-lines-v1:sha256:8928ccb9c60bd909775f025ef6b26e7b7f6371f068b00c1a85abe710d57cb396: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"
        },
        {
          "resource": "repo://libs/langchain_v1/pyproject.toml#L24-L24",
          "version": "repo-lines-v1:sha256:53cfa819889bdf01f005c2fb74a4d3a902cdb4ca06b0d8dca027c701510c7201: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"
        }
      ]
    },
    {
      "id": "claim_ca180050b7ca46b7ae59de05eefc84b7",
      "statement": "OpenAI Decisions API integration is implemented in libs/partners/openai/langchain_openai/decisions/, providing structured output marshaling and response routing for deterministic model outputs.",
      "evidence": [
        {
          "resource": "repo://libs/partners/openai/langchain_openai/decisions/base.py",
          "version": "repo-file-v1:sha256:430531c926f369c474d95856b96e19e137dca1f01a9228c2f4e54be3ecd7bfed"
        }
      ]
    }
  ],
  "verification": {
    "by": "openwiki/0.5.0",
    "at": "2026-10-22T08:26:56.662Z"
  }
}