sinapsis-anthropic 0.2.0

Sinapsis templates for Anthropic Claude models.

Sinapsis Anthropic

Sinapsis templates for Anthropic Claude text, multimodal, and MCP workflows.

🐍 Installation • 🚀 Features • 📚 Usage example • 📙 Documentation • 🔍 License

The sinapsis-anthropic package provides packet-native templates for running Claude models with LLMConversationPacket, including standard text generation, multimodal inputs, and MCP tool flows.

🐍 Installation

Install using your preferred package manager. We strongly recommend using uv.

Install the base package:

uv pip install sinapsis-anthropic --extra-index-url https://pypi.sinapsis.tech

Or with raw pip:

pip install sinapsis-anthropic --extra-index-url https://pypi.sinapsis.tech

[!IMPORTANT] Anthropic templates require an API key. Set it before running your agent:

export ANTHROPIC_API_KEY="your-api-key"

[!IMPORTANT] AnthropicMultiModal requires the optional multimodal extra:

uv pip install sinapsis-anthropic[multimodal] --extra-index-url https://pypi.sinapsis.tech

Or install all optional dependencies:

uv pip install sinapsis-anthropic[all] --extra-index-url https://pypi.sinapsis.tech

🚀 Features

Templates Supported

• AnthropicTextGeneration: Standard Claude text generation on top of LLMConversationPacket.
• AnthropicMultiModal: Multimodal completion that reads image packets from container.images.
• AnthropicWithMCP: Packet-native MCP orchestration using LLMConversationPacket.tools, tool_calls, and tool_results.

🧩 Common Attributes

• init_args (LLMInitArgs, required): Anthropic model selection arguments.
  • llm_model_name (str, required): Claude model identifier.
• completion_args (AnthropicCompletionArgs, required): Request-time generation parameters such as max_tokens, temperature, top_p, top_k, service_tier, and stop_sequences.
• extended_thinking (AnthropicThinkingArgs, optional): Extended-thinking configuration.
• web_search (bool, optional): Enables Anthropic's built-in web search tool for supported models.
• cleanup_response (bool, optional): Whether to clean leaked transport or template markers from the final response.
• reasoning_start_tag / reasoning_end_tag (str | None, optional): Optional tags used to extract reasoning into the packet reasoning field and keep the final answer in response.

[!TIP] Use sinapsis info --all-template-names to inspect the templates installed in your environment.

📚 Usage example

The following configuration runs one Claude text-generation step from an LLMConversationPacket.

Config

agent:
  name: my_claude_agent
  description: Agent with support for text-to-text conversational chat using Anthropic Claude models

templates:
  - template_name: InputTemplate
    class_name: InputTemplate
    attributes: {}

  - template_name: LLMConversationInput
    class_name: LLMConversationInput
    template_input: InputTemplate
    attributes:
      prompt: "Give three short tips for staying organized during a busy week."
      system_prompt: "You are a helpful assistant."

  - template_name: AnthropicTextGeneration
    class_name: AnthropicTextGeneration
    template_input: LLMConversationInput
    attributes:
      init_args:
        llm_model_name: "claude-sonnet-4-6"
      completion_args:
        max_tokens: 1024
        temperature: 0.7
        service_tier: standard_only
      extended_thinking:
        type: disabled
      web_search: false

To run a config:

sinapsis run path_to_config.yaml

📙 Documentation

Documentation for this and other sinapsis packages is available on the sinapsis website

Tutorials for different projects within sinapsis are available at sinapsis tutorials page

🔍 License

This project is licensed under the AGPLv3 license, which encourages open collaboration and sharing. For more details, please refer to the LICENSE file.

For commercial use, please refer to our official Sinapsis website for information on obtaining a commercial license.