dapr
/
dapr-agents
mirror of https://github.com/dapr/dapr-agents.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity

Added brief MCP support explanation (#116) 

* Added brief MCP support explanation

Signed-off-by: Bilgin Ibryam <bibryam@gmail.com>

* Update docs/concepts/agents.md

Co-authored-by: Casper Nielsen <scni@novonordisk.com>
Signed-off-by: Bilgin Ibryam <bibryam@gmail.com>

---------

Signed-off-by: Bilgin Ibryam <bibryam@gmail.com>
Co-authored-by: Casper Nielsen <scni@novonordisk.com>
Co-authored-by: Yaron Schneider <schneider.yaron@live.com>
This commit is contained in:
Bilgin Ibryam 2025-05-12 16:44:19 +01:00 committed by GitHub
parent 41faa4f5b7
commit 8741289e7d
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
2 changed files with 12 additions and 6 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

16
docs/concepts/agents.md
View File

@ -18,23 +18,29 @@ Agents in Dapr Agents leverage structured output capabilities, such as [OpenAI
Agents dynamically select the appropriate tool for a given task, using LLMs to analyze requirements and choose the best action. This is supported directly through LLM parametric knowledge and enhanced by [Function Calling](https://platform.openai.com/docs/guides/function-calling), ensuring tools are invoked efficiently and accurately.
### 4. Memory
### 4. MCP Support
Dapr Agents includes built-in support for the [Model Context Protocol (MCP)](https://modelcontextprotocol.io/), enabling agents to dynamically discover and invoke external tools through a standardized interface. Using the provided MCPClient, agents can connect to MCP servers via two transport options: stdio for local development and sse for remote or distributed environments.
Once connected, the MCP client fetches all available tools from the server and prepares them for immediate use within the agent’s toolset. This allows agents to incorporate capabilities exposed by external processes—such as local Python scripts or remote services without hardcoding or preloading them. Agents can invoke these tools at runtime, expanding their behavior based on what’s offered by the active MCP server.
### 5. Memory
Agents retain context across interactions, enhancing their ability to provide coherent and adaptive responses. Memory options range from simple in-memory lists for managing chat history to vector databases for semantic search and retrieval. Dapr Agents also integrates with [Dapr state stores](https://docs.dapr.io/developing-applications/building-blocks/state-management/howto-get-save-state/), enabling scalable and persistent memory for advanced use cases.
### 5. Prompt Flexibility
### 6. Prompt Flexibility
Dapr Agents supports flexible prompt templates to shape agent behavior and reasoning. Users can define placeholders within prompts, enabling dynamic input of context for inference calls. By leveraging prompt formatting with [Jinja templates](https://jinja.palletsprojects.com/en/stable/templates/), users can include loops, conditions, and variables, providing precise control over the structure and content of prompts. This flexibility ensures that LLM responses are tailored to the task at hand, offering modularity and adaptability for diverse use cases.
### 6. Agent Services
### 7. Agent Services
Agents are exposed as independent services using [FastAPI and Dapr applications](https://docs.dapr.io/developing-applications/sdks/python/python-sdk-extensions/python-fastapi/). This modular approach separates the agent’s logic from its service layer, enabling seamless reuse, deployment, and integration into multi-agent systems.
### 7. Message-Driven Communication
### 8. Message-Driven Communication
Agents collaborate through [Pub/Sub messaging](https://docs.dapr.io/developing-applications/building-blocks/pubsub/pubsub-overview/), enabling event-driven communication and task distribution. This message-driven architecture allows agents to work asynchronously, share updates, and respond to real-time events, ensuring effective collaboration in distributed systems.
### 8. Workflow Orchestration
### 9. Workflow Orchestration
Dapr Agents supports both deterministic and event-driven workflows to manage multi-agent systems via [Dapr Workflows](https://docs.dapr.io/developing-applications/building-blocks/workflow/workflow-overview/). Deterministic workflows provide clear, repeatable processes, while event-driven workflows allow for dynamic, adaptive collaboration between agents in centralized or decentralized architectures.

2
docs/index.md
View File

@ -26,7 +26,7 @@ Dapr Agents builds on top of Dapr's Workflow API, which under the hood represent
### Data-Centric AI Agents
With built-in connectivity to over 50 enterprise data sources, Dapr Agents efficiently handles structured and unstructured data. From basic [PDF extraction](./concepts/arxiv_fetcher.md) to large-scale database interactions, it enables seamless data-driven AI workflows with minimal code changes. Dapr's [bindings](https://docs.dapr.io/reference/components-reference/supported-bindings/) and [state stores](https://docs.dapr.io/reference/components-reference/supported-state-stores/) provide access to a large number of data sources that can be used to ingest data to an agent.
With built-in connectivity to over 50 enterprise data sources, Dapr Agents efficiently handles structured and unstructured data. From basic [PDF extraction](./concepts/arxiv_fetcher.md) to large-scale database interactions, it enables seamless data-driven AI workflows with minimal code changes. Dapr's [bindings](https://docs.dapr.io/reference/components-reference/supported-bindings/) and [state stores](https://docs.dapr.io/reference/components-reference/supported-state-stores/), along with [MCP](https://modelcontextprotocol.io/) support, provide access to a large number of data sources that can be used to ingest data to an agent.
### Accelerated Development