PyPI - solace-agent-mesh - Versions diffs - 1.0.9__py3-none-any.whl → 1.3.0__py3-none-any.whl - Mend

solace-agent-mesh 1.0.9py3-none-any.whl → 1.3.0py3-none-any.whl

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.

Potentially problematic release.

This version of solace-agent-mesh might be problematic. Click here for more details.

Files changed (220) hide show

solace_agent_mesh/agent/utils/utils_llm.txt CHANGED Viewed

@@ -2,152 +2,137 @@
 The `utils` directory provides a collection of helper modules designed to support the core functionality of the agent. These utilities encapsulate common, reusable logic for tasks such as artifact management (saving, loading, schema inference), configuration parsing, and safe interaction with the ADK's invocation context.
 ## Files Overview
-- `__init__.py` - An empty file that marks the directory as a Python package.
-- `artifact_helpers.py` - Provides functions for creating, retrieving, and managing artifacts and their associated metadata.
-- `config_parser.py` - Contains helpers for parsing and validating agent and app configurations.
-- `context_helpers.py` - Offers utility functions for safely extracting data from ADK callback and invocation contexts.
+- `__init__.py` - Empty package marker file
+- `artifact_helpers.py` - Comprehensive artifact management functions including save/load operations, metadata handling, and schema inference
+- `config_parser.py` - Configuration parsing utilities for resolving instruction providers
+- `context_helpers.py` - Safe utilities for extracting data from ADK callback and invocation contexts
 ## Developer API Reference
 ### artifact_helpers.py
-**Purpose:** This module offers a comprehensive set of asynchronous functions for interacting with an artifact storage service. It handles saving artifacts with automatically generated metadata, inferring schemas for common data types, loading artifact content or just metadata, and listing available artifacts.
-**Import:** `from src.solace_agent_mesh.agent.utils.artifact_helpers import is_filename_safe, ensure_correct_extension, save_artifact_with_metadata, load_artifact_content_or_metadata, get_latest_artifact_version, get_artifact_info_list, format_metadata_for_llm, decode_and_get_bytes`
+**Purpose:** Comprehensive artifact management with automatic metadata generation, schema inference, and async operations
+**Import:** `from solace_agent_mesh.agent.utils.artifact_helpers import save_artifact_with_metadata, load_artifact_content_or_metadata, get_artifact_info_list, is_filename_safe, ensure_correct_extension`
 **Functions:**
-- `is_filename_safe(filename: str) -> bool` - Checks if a filename is safe for artifact creation (e.g., no path traversal).
-- `ensure_correct_extension(filename_from_llm: str, desired_extension: str) -> str` - Corrects or adds the desired file extension to a filename.
-- `save_artifact_with_metadata(artifact_service: BaseArtifactService, app_name: str, user_id: str, session_id: str, filename: str, content_bytes: bytes, mime_type: str, metadata_dict: Dict[str, Any], timestamp: datetime.datetime, explicit_schema: Optional[Dict] = None, schema_inference_depth: int = 2, schema_max_keys: int = 20, tool_context: Optional["ToolContext"] = None) -> Dict[str, Any]` - Asynchronously saves a data artifact and a corresponding metadata artifact, with optional schema inference.
-- `format_metadata_for_llm(metadata: Dict[str, Any]) -> str` - Formats an artifact's metadata dictionary into a human-readable, LLM-friendly string.
-- `decode_and_get_bytes(content_str: str, mime_type: str, log_identifier: str) -> Tuple[bytes, str]` - Decodes a string into bytes, handling base64 for binary MIME types and falling back to UTF-8 for text.
-- `get_latest_artifact_version(artifact_service: BaseArtifactService, app_name: str, user_id: str, session_id: str, filename: str) -> Optional[int]` - Asynchronously retrieves the latest version number for a given artifact.
-- `get_artifact_info_list(artifact_service: BaseArtifactService, app_name: str, user_id: str, session_id: str) -> List[ArtifactInfo]` - Asynchronously retrieves a list of detailed `ArtifactInfo` objects for all artifacts in the current context.
-- `load_artifact_content_or_metadata(artifact_service: BaseArtifactService, app_name: str, user_id: str, session_id: str, filename: str, version: Union[int, str], load_metadata_only: bool = False, return_raw_bytes: bool = False, max_content_length: Optional[int] = None, component: Optional[Any] = None, encoding: str = "utf-8", error_handling: str = "strict") -> Dict[str, Any]` - Asynchronously loads an artifact's content or just its metadata for a specific version.
+- `is_filename_safe(filename: str) -> bool` - Validates filename safety (no path traversal, separators, or reserved names)
+- `ensure_correct_extension(filename_from_llm: str, desired_extension: str) -> str` - Ensures filename has correct extension
+- `format_artifact_uri(app_name: str, user_id: str, session_id: str, filename: str, version: Union[int, str]) -> str` - Formats components into standard artifact:// URI
+- `parse_artifact_uri(uri: str) -> Dict[str, Any]` - Parses artifact:// URI into constituent parts
+- `save_artifact_with_metadata(artifact_service: BaseArtifactService, app_name: str, user_id: str, session_id: str, filename: str, content_bytes: bytes, mime_type: str, metadata_dict: Dict[str, Any], timestamp: datetime, explicit_schema: Optional[Dict] = None, schema_inference_depth: int = 2, schema_max_keys: int = 20, tool_context: Optional["ToolContext"] = None) -> Dict[str, Any]` - Saves artifact with auto-generated metadata and schema inference
+- `load_artifact_content_or_metadata(artifact_service: BaseArtifactService, app_name: str, user_id: str, session_id: str, filename: str, version: Union[int, str], load_metadata_only: bool = False, return_raw_bytes: bool = False, max_content_length: Optional[int] = None, component: Optional[Any] = None, log_identifier_prefix: str = "[ArtifactHelper:load]", encoding: str = "utf-8", error_handling: str = "strict") -> Dict[str, Any]` - Loads artifact content or metadata with flexible options
+- `get_artifact_info_list(artifact_service: BaseArtifactService, app_name: str, user_id: str, session_id: str) -> List[ArtifactInfo]` - Retrieves detailed info for all artifacts
+- `get_latest_artifact_version(artifact_service: BaseArtifactService, app_name: str, user_id: str, session_id: str, filename: str) -> Optional[int]` - Gets latest version number for an artifact
+- `format_metadata_for_llm(metadata: Dict[str, Any]) -> str` - Formats metadata into LLM-friendly text
+- `decode_and_get_bytes(content_str: str, mime_type: str, log_identifier: str) -> Tuple[bytes, str]` - Decodes content based on MIME type (base64 for binary, UTF-8 for text)
+- `generate_artifact_metadata_summary(component: "SamAgentComponent", artifact_identifiers: List[Dict[str, Any]], user_id: str, session_id: str, app_name: str, header_text: Optional[str] = None) -> str` - Generates YAML summary of multiple artifacts' metadata
 **Constants/Variables:**
-- `METADATA_SUFFIX: str` - The suffix used for metadata files, `".metadata.json"`.
-- `DEFAULT_SCHEMA_MAX_KEYS: int` - The default maximum number of keys to inspect when inferring a schema for a dictionary (default: 20).
+- `METADATA_SUFFIX: str` - Suffix for metadata files (".metadata.json")
+- `DEFAULT_SCHEMA_MAX_KEYS: int` - Default max keys for schema inference (20)
 **Usage Examples:**
 ```python
 import asyncio
-import datetime
-from typing import Dict, Any, Optional, List, Union
-from google.adk.artifacts import BaseArtifactService, MemoryArtifactService
-from src.solace_agent_mesh.agent.utils.artifact_helpers import (
+from datetime import datetime, timezone
+from solace_agent_mesh.agent.utils.artifact_helpers import (
     save_artifact_with_metadata,
     load_artifact_content_or_metadata,
     get_artifact_info_list,
-    ensure_correct_extension
+    ensure_correct_extension,
+    format_artifact_uri,
+    parse_artifact_uri
 )
-# Assume 'service' is an instance of a BaseArtifactService implementation
-service = MemoryArtifactService()
-app_name = "my_app"
-user_id = "user-1"
-session_id = "session-1"
-async def manage_artifacts():
-    # 1. Ensure the filename from an LLM has the correct extension
-    safe_filename = ensure_correct_extension("quarterly_report", "csv") # -> "quarterly_report.csv"
-    # 2. Save an artifact with some metadata
-    csv_content = b"id,name\n1,Alice\n2,Bob"
-    save_result = await save_artifact_with_metadata(
+async def artifact_example():
+    # Ensure safe filename
+    safe_name = ensure_correct_extension("report", "csv")  # -> "report.csv"
+    # Save artifact with metadata
+    csv_data = b"name,age\nAlice,30\nBob,25"
+    result = await save_artifact_with_metadata(
         artifact_service=service,
-        app_name=app_name,
-        user_id=user_id,
-        session_id=session_id,
-        filename=safe_filename,
-        content_bytes=csv_content,
+        app_name="my_app",
+        user_id="user123",
+        session_id="session456",
+        filename=safe_name,
+        content_bytes=csv_data,
         mime_type="text/csv",
-        metadata_dict={"source": "manual_upload", "description": "User sales data"},
-        timestamp=datetime.datetime.now(datetime.timezone.utc)
+        metadata_dict={"source": "user_upload", "description": "Employee data"},
+        timestamp=datetime.now(timezone.utc)
     )
-    print(f"Save result: {save_result}")
-    # 3. Load the content of the artifact we just saved
-    loaded_artifact = await load_artifact_content_or_metadata(
+    # Load artifact content
+    loaded = await load_artifact_content_or_metadata(
         artifact_service=service,
-        app_name=app_name,
-        user_id=user_id,
-        session_id=session_id,
-        filename=safe_filename,
+        app_name="my_app",
+        user_id="user123",
+        session_id="session456",
+        filename=safe_name,
         version="latest"
     )
-    if loaded_artifact.get("status") == "success":
-        print(f"Loaded content: {loaded_artifact.get('content')}")
-    # 4. List all available artifacts
-    all_artifacts = await get_artifact_info_list(
+    # Work with artifact URIs
+    uri = format_artifact_uri("my_app", "user123", "session456", "report.csv", 1)
+    # Returns: "artifact://my_app/user123/session456/report.csv?version=1"
+    parsed = parse_artifact_uri(uri)
+    # Returns: {"app_name": "my_app", "user_id": "user123", ...}
+    # List all artifacts
+    artifacts = await get_artifact_info_list(
         artifact_service=service,
-        app_name=app_name,
-        user_id=user_id,
-        session_id=session_id
+        app_name="my_app",
+        user_id="user123",
+        session_id="session456"
     )
-    for artifact in all_artifacts:
-        print(f"Found artifact: {artifact.filename} (v{artifact.version})")
-# To run the async example
-# asyncio.run(manage_artifacts())
 ```
 ### config_parser.py
-**Purpose:** This module provides a utility to resolve configuration values that can be either a static string or a dynamic, callable "invoke" block. This is primarily used for parsing the agent's `instruction` configuration.
-**Import:** `from src.solace_agent_mesh.agent.utils.config_parser import resolve_instruction_provider`
+**Purpose:** Resolves configuration values that can be static strings or dynamic callable providers
+**Import:** `from solace_agent_mesh.agent.utils.config_parser import resolve_instruction_provider, InstructionProvider`
 **Functions:**
-- `resolve_instruction_provider(component, config_value: Any) -> Union[str, Callable[[ReadonlyContext], str]]` - Resolves an instruction from a configuration value. If the value is a string, it's returned directly. If it's a dictionary with an "invoke" key that resolves to a callable, the callable is returned.
+- `resolve_instruction_provider(component, config_value: Any) -> Union[str, InstructionProvider]` - Resolves instruction config from string or invoke block
 **Usage Examples:**
 ```python
-# Assume 'my_agent_component' is an instance of an agent component
-# and it has a method 'get_config'
+from solace_agent_mesh.agent.utils.config_parser import resolve_instruction_provider
-# Example 1: Config value is a simple string
-config_str = "You are a helpful assistant."
-instruction = resolve_instruction_provider(my_agent_component, config_str)
-# instruction is "You are a helpful assistant."
+# Static string instruction
+instruction = resolve_instruction_provider(component, "You are a helpful assistant.")
+# Returns: "You are a helpful assistant."
-# Example 2: Config value is a callable (e.g., from a YAML 'invoke' block)
-def instruction_provider(context):
-    return f"Assistant for user {context.user_id}"
+# Dynamic instruction provider (from YAML invoke block)
+def dynamic_instruction(context):
+    return f"Assistant for {context.user_id}"
-instruction_func = resolve_instruction_provider(my_agent_component, instruction_provider)
-# instruction_func is the instruction_provider function
+instruction_func = resolve_instruction_provider(component, dynamic_instruction)
+# Returns: the callable function
 ```
 ### context_helpers.py
-**Purpose:** This module provides safe and stable helper functions to extract information from the ADK's `CallbackContext` and other invocation context objects, abstracting away internal attribute access.
-**Import:** `from src.solace_agent_mesh.agent.utils.context_helpers import get_session_from_callback_context, get_original_session_id`
+**Purpose:** Safe utilities for extracting information from ADK contexts
+**Import:** `from solace_agent_mesh.agent.utils.context_helpers import get_session_from_callback_context, get_original_session_id`
 **Functions:**
-- `get_session_from_callback_context(callback_context: CallbackContext) -> Session` - Safely retrieves the persistent `Session` object from a `CallbackContext`.
-- `get_original_session_id(invocation_context: Any) -> str` - Extracts the base session ID from a context, stripping any suffixes added after a colon (e.g., "session123:tool456" -> "session123").
+- `get_session_from_callback_context(callback_context: CallbackContext) -> Session` - Safely extracts Session object from CallbackContext
+- `get_original_session_id(invocation_context: Any) -> str` - Extracts base session ID, removing any colon-separated suffixes
 **Usage Examples:**
 ```python
-from google.adk.agents.callback_context import CallbackContext
-from google.adk.sessions import Session, SessionId
-from src.solace_agent_mesh.agent.utils.context_helpers import get_session_from_callback_context, get_original_session_id
-# Mock context objects for demonstration
-class MockInvocationContext:
-    def __init__(self, session_id_str: str):
-        self.session = Session(id=SessionId(session_id_str))
-# 1. Get the full session object from a callback context
-# In a real tool, 'callback_context' is provided by the ADK
-mock_callback_context = CallbackContext(_invocation_context=MockInvocationContext("session123:tool456"))
-session_obj = get_session_from_callback_context(mock_callback_context)
-print(f"Full Session ID: {session_obj.id}") # -> "session123:tool456"
-# 2. Get the original session ID from an invocation context
-# In a real tool, this comes from 'tool_context._invocation_context'
-mock_inv_context = MockInvocationContext("session123:tool456")
-original_id = get_original_session_id(mock_inv_context)
-print(f"Original Session ID: {original_id}") # -> "session123"
-# Works with simple IDs too
-mock_inv_context_simple = MockInvocationContext("session789")
-original_id_simple = get_original_session_id(mock_inv_context_simple)
-print(f"Original Session ID (simple): {original_id_simple}") # -> "session789
+from solace_agent_mesh.agent.utils.context_helpers import (
+    get_session_from_callback_context,
+    get_original_session_id
+)
+# In a tool function with callback_context
+def my_tool(callback_context):
+    # Get full session object
+    session = get_session_from_callback_context(callback_context)
+    # Get original session ID (strips suffixes after colon)
+    original_id = get_original_session_id(tool_context._invocation_context)
+    # "session123:tool456" -> "session123"
+```
+# content_hash: b3aa9a05daf2be266a8a43107328da63d05222408df1b57e46af1c75415d5b7d

solace_agent_mesh/assets/docs/404.html CHANGED Viewed

@@ -4,13 +4,13 @@
 <meta charset="UTF-8">
 <meta name="generator" content="Docusaurus v3.8.1">
 <title data-rh="true">Page Not Found | Solace Agent Mesh</title><meta data-rh="true" name="viewport" content="width=device-width,initial-scale=1"><meta data-rh="true" name="twitter:card" content="summary_large_image"><meta data-rh="true" property="og:image" content="https://solacelabs.github.io/solace-agent-mesh/img/logo.png"><meta data-rh="true" name="twitter:image" content="https://solacelabs.github.io/solace-agent-mesh/img/logo.png"><meta data-rh="true" property="og:url" content="https://solacelabs.github.io/solace-agent-mesh/404.html"><meta data-rh="true" property="og:locale" content="en"><meta data-rh="true" name="docusaurus_locale" content="en"><meta data-rh="true" name="docusaurus_tag" content="default"><meta data-rh="true" name="docsearch:language" content="en"><meta data-rh="true" name="docsearch:docusaurus_tag" content="default"><meta data-rh="true" property="og:title" content="Page Not Found | Solace Agent Mesh"><link data-rh="true" rel="icon" href="/solace-agent-mesh/img/logo.png"><link data-rh="true" rel="canonical" href="https://solacelabs.github.io/solace-agent-mesh/404.html"><link data-rh="true" rel="alternate" href="https://solacelabs.github.io/solace-agent-mesh/404.html" hreflang="en"><link data-rh="true" rel="alternate" href="https://solacelabs.github.io/solace-agent-mesh/404.html" hreflang="x-default"><link rel="stylesheet" href="/solace-agent-mesh/assets/css/styles.906a1503.css">
-<script src="/solace-agent-mesh/assets/js/runtime~main.05d19492.js" defer="defer"></script>
-<script src="/solace-agent-mesh/assets/js/main.3d0e7879.js" defer="defer"></script>
+<script src="/solace-agent-mesh/assets/js/runtime~main.458efb1d.js" defer="defer"></script>
+<script src="/solace-agent-mesh/assets/js/main.08d30374.js" defer="defer"></script>
 </head>
 <body class="navigation-with-keyboard">
 <svg xmlns="http://www.w3.org/2000/svg" style="display: none;"><defs>
 <symbol id="theme-svg-external-link" viewBox="0 0 24 24"><path fill="currentColor" d="M21 13v10h-21v-19h12v2h-10v15h17v-8h2zm3-12h-10.988l4.035 4-6.977 7.07 2.828 2.828 6.977-7.07 4.125 4.172v-11z"/></symbol>
 </defs></svg>
-<script>!function(){var t=function(){try{return new URLSearchParams(window.location.search).get("docusaurus-theme")}catch(t){}}()||function(){try{return window.localStorage.getItem("theme")}catch(t){}}();document.documentElement.setAttribute("data-theme",t||(window.matchMedia("(prefers-color-scheme: dark)").matches?"dark":"light")),document.documentElement.setAttribute("data-theme-choice",t||"system")}(),function(){try{const c=new URLSearchParams(window.location.search).entries();for(var[t,e]of c)if(t.startsWith("docusaurus-data-")){var a=t.replace("docusaurus-data-","data-");document.documentElement.setAttribute(a,e)}}catch(t){}}()</script><div id="__docusaurus"><link rel="preload" as="image" href="/solace-agent-mesh/img/logo.png"><link rel="preload" as="image" href="/solace-agent-mesh/img/solace-logo.png"><div role="region" aria-label="Skip to main content"><a class="skipToContent_fXgn" href="#__docusaurus_skipToContent_fallback">Skip to main content</a></div><nav aria-label="Main" class="theme-layout-navbar navbar navbar--fixed-top"><div class="navbar__inner"><div class="theme-layout-navbar-left navbar__items"><button aria-label="Toggle navigation bar" aria-expanded="false" class="navbar__toggle clean-btn" type="button"><svg width="30" height="30" viewBox="0 0 30 30" aria-hidden="true"><path stroke="currentColor" stroke-linecap="round" stroke-miterlimit="10" stroke-width="2" d="M4 7h22M4 15h22M4 23h22"></path></svg></button><a class="navbar__brand" href="/solace-agent-mesh/docs/documentation/getting-started/introduction"><div class="navbar__logo"><img src="/solace-agent-mesh/img/logo.png" alt="Solace Agent Mesh Logo" class="themedComponent_mlkZ themedComponent--light_NVdE"><img src="/solace-agent-mesh/img/logo.png" alt="Solace Agent Mesh Logo" class="themedComponent_mlkZ themedComponent--dark_xIcU"></div><b class="navbar__title text--truncate">Solace Agent Mesh</b></a><a class="navbar__item navbar__link" href="/solace-agent-mesh/docs/documentation/getting-started/introduction">Documentation</a></div><div class="theme-layout-navbar-right navbar__items navbar__items--right"><a href="https://github.com/SolaceLabs/solace-agent-mesh/" target="_blank" rel="noopener noreferrer" class="navbar__item navbar__link">GitHub<svg width="13.5" height="13.5" aria-hidden="true" class="iconExternalLink_nPIU"><use href="#theme-svg-external-link"></use></svg></a><div class="toggle_vylO colorModeToggle_DEke"><button class="clean-btn toggleButton_gllP toggleButtonDisabled_aARS" type="button" disabled="" title="system mode" aria-label="Switch between dark and light mode (currently system mode)"><svg viewBox="0 0 24 24" width="24" height="24" aria-hidden="true" class="toggleIcon_g3eP lightToggleIcon_pyhR"><path fill="currentColor" d="M12,9c1.65,0,3,1.35,3,3s-1.35,3-3,3s-3-1.35-3-3S10.35,9,12,9 M12,7c-2.76,0-5,2.24-5,5s2.24,5,5,5s5-2.24,5-5 S14.76,7,12,7L12,7z M2,13l2,0c0.55,0,1-0.45,1-1s-0.45-1-1-1l-2,0c-0.55,0-1,0.45-1,1S1.45,13,2,13z M20,13l2,0c0.55,0,1-0.45,1-1 s-0.45-1-1-1l-2,0c-0.55,0-1,0.45-1,1S19.45,13,20,13z M11,2v2c0,0.55,0.45,1,1,1s1-0.45,1-1V2c0-0.55-0.45-1-1-1S11,1.45,11,2z M11,20v2c0,0.55,0.45,1,1,1s1-0.45,1-1v-2c0-0.55-0.45-1-1-1C11.45,19,11,19.45,11,20z M5.99,4.58c-0.39-0.39-1.03-0.39-1.41,0 c-0.39,0.39-0.39,1.03,0,1.41l1.06,1.06c0.39,0.39,1.03,0.39,1.41,0s0.39-1.03,0-1.41L5.99,4.58z M18.36,16.95 c-0.39-0.39-1.03-0.39-1.41,0c-0.39,0.39-0.39,1.03,0,1.41l1.06,1.06c0.39,0.39,1.03,0.39,1.41,0c0.39-0.39,0.39-1.03,0-1.41 L18.36,16.95z M19.42,5.99c0.39-0.39,0.39-1.03,0-1.41c-0.39-0.39-1.03-0.39-1.41,0l-1.06,1.06c-0.39,0.39-0.39,1.03,0,1.41 s1.03,0.39,1.41,0L19.42,5.99z M7.05,18.36c0.39-0.39,0.39-1.03,0-1.41c-0.39-0.39-1.03-0.39-1.41,0l-1.06,1.06 c-0.39,0.39-0.39,1.03,0,1.41s1.03,0.39,1.41,0L7.05,18.36z"></path></svg><svg viewBox="0 0 24 24" width="24" height="24" aria-hidden="true" class="toggleIcon_g3eP darkToggleIcon_wfgR"><path fill="currentColor" d="M9.37,5.51C9.19,6.15,9.1,6.82,9.1,7.5c0,4.08,3.32,7.4,7.4,7.4c0.68,0,1.35-0.09,1.99-0.27C17.45,17.19,14.93,19,12,19 c-3.86,0-7-3.14-7-7C5,9.07,6.81,6.55,9.37,5.51z M12,3c-4.97,0-9,4.03-9,9s4.03,9,9,9s9-4.03,9-9c0-0.46-0.04-0.92-0.1-1.36 c-0.98,1.37-2.58,2.26-4.4,2.26c-2.98,0-5.4-2.42-5.4-5.4c0-1.81,0.89-3.42,2.26-4.4C12.92,3.04,12.46,3,12,3L12,3z"></path></svg><svg viewBox="0 0 24 24" width="24" height="24" aria-hidden="true" class="toggleIcon_g3eP systemToggleIcon_QzmC"><path fill="currentColor" d="m12 21c4.971 0 9-4.029 9-9s-4.029-9-9-9-9 4.029-9 9 4.029 9 9 9zm4.95-13.95c1.313 1.313 2.05 3.093 2.05 4.95s-0.738 3.637-2.05 4.95c-1.313 1.313-3.093 2.05-4.95 2.05v-14c1.857 0 3.637 0.737 4.95 2.05z"></path></svg></button></div><div class="navbarSearchContainer_Bca1"><div class="navbar__search"><span aria-label="expand searchbar" role="button" class="search-icon" tabindex="0"></span><input id="search_input_react" type="search" placeholder="Loading..." aria-label="Search" class="navbar__search-input search-bar" disabled=""></div></div></div></div><div role="presentation" class="navbar-sidebar__backdrop"></div></nav><div id="__docusaurus_skipToContent_fallback" class="theme-layout-main main-wrapper mainWrapper_z2l0"><main class="container margin-vert--xl"><div class="row"><div class="col col--6 col--offset-3"><h1 class="hero__title">Page Not Found</h1><p>We could not find what you were looking for.</p><p>Please contact the owner of the site that linked you to the original URL and let them know their link is broken.</p></div></div></main></div><footer class="theme-layout-footer footer footer--dark"><div class="container container-fluid"><div class="row footer__links"><div class="theme-layout-footer-column col footer__col"><div class="footer__title">Solace Agent Mesh</div><ul class="footer__items clean-list"><li class="footer__item"><a class="footer__link-item" href="/solace-agent-mesh/docs/documentation/getting-started/introduction">Documentation</a></li><li class="footer__item"><a href="https://github.com/SolaceLabs/solace-agent-mesh/" target="_blank" rel="noopener noreferrer" class="footer__link-item">GitHub<svg width="13.5" height="13.5" aria-hidden="true" class="iconExternalLink_nPIU"><use href="#theme-svg-external-link"></use></svg></a></li><li class="footer__item"><a href="https://github.com/SolaceLabs/solace-agent-mesh-core-plugins/" target="_blank" rel="noopener noreferrer" class="footer__link-item">Official Plugins<svg width="13.5" height="13.5" aria-hidden="true" class="iconExternalLink_nPIU"><use href="#theme-svg-external-link"></use></svg></a></li></ul></div><div class="theme-layout-footer-column col footer__col"><div class="footer__title">Company</div><ul class="footer__items clean-list"><li class="footer__item"><a href="https://solace.com/products/" target="_blank" rel="noopener noreferrer" class="footer__link-item">Products<svg width="13.5" height="13.5" aria-hidden="true" class="iconExternalLink_nPIU"><use href="#theme-svg-external-link"></use></svg></a></li><li class="footer__item"><a href="https://solace.com/contact/" target="_blank" rel="noopener noreferrer" class="footer__link-item">Contact<svg width="13.5" height="13.5" aria-hidden="true" class="iconExternalLink_nPIU"><use href="#theme-svg-external-link"></use></svg></a></li><li class="footer__item"><a href="https://solace.com/support/" target="_blank" rel="noopener noreferrer" class="footer__link-item">Support<svg width="13.5" height="13.5" aria-hidden="true" class="iconExternalLink_nPIU"><use href="#theme-svg-external-link"></use></svg></a></li><li class="footer__item"><a href="https://solace.com/legal/" target="_blank" rel="noopener noreferrer" class="footer__link-item">Privacy and Legal<svg width="13.5" height="13.5" aria-hidden="true" class="iconExternalLink_nPIU"><use href="#theme-svg-external-link"></use></svg></a></li></ul></div><div class="theme-layout-footer-column col footer__col"><div class="footer__title">Community</div><ul class="footer__items clean-list"><li class="footer__item"><a href="https://www.linkedin.com/company/solacedotcom/" target="_blank" rel="noopener noreferrer" class="footer__link-item">LinkedIn<svg width="13.5" height="13.5" aria-hidden="true" class="iconExternalLink_nPIU"><use href="#theme-svg-external-link"></use></svg></a></li><li class="footer__item"><a href="https://github.com/SolaceLabs" target="_blank" rel="noopener noreferrer" class="footer__link-item">GitHub<svg width="13.5" height="13.5" aria-hidden="true" class="iconExternalLink_nPIU"><use href="#theme-svg-external-link"></use></svg></a></li><li class="footer__item"><a href="https://www.youtube.com/SolaceSystems" target="_blank" rel="noopener noreferrer" class="footer__link-item">YouTube<svg width="13.5" height="13.5" aria-hidden="true" class="iconExternalLink_nPIU"><use href="#theme-svg-external-link"></use></svg></a></li><li class="footer__item"><a href="https://twitter.com/solacedotcom" target="_blank" rel="noopener noreferrer" class="footer__link-item">X<svg width="13.5" height="13.5" aria-hidden="true" class="iconExternalLink_nPIU"><use href="#theme-svg-external-link"></use></svg></a></li></ul></div></div><div class="footer__bottom text--center"><div class="margin-bottom--sm"><img src="/solace-agent-mesh/img/solace-logo.png" alt="Solace Logo" class="footer__logo themedComponent_mlkZ themedComponent--light_NVdE" width="10%" height="10%"><img src="/solace-agent-mesh/img/solace-logo.png" alt="Solace Logo" class="footer__logo themedComponent_mlkZ themedComponent--dark_xIcU" width="10%" height="10%"></div><div class="footer__copyright">Solace Agent Mesh. Copyright © 2025 Solace. Version: 1.0.9</div></div></div></footer></div>
+<script>!function(){var t=function(){try{return new URLSearchParams(window.location.search).get("docusaurus-theme")}catch(t){}}()||function(){try{return window.localStorage.getItem("theme")}catch(t){}}();document.documentElement.setAttribute("data-theme",t||(window.matchMedia("(prefers-color-scheme: dark)").matches?"dark":"light")),document.documentElement.setAttribute("data-theme-choice",t||"system")}(),function(){try{const c=new URLSearchParams(window.location.search).entries();for(var[t,e]of c)if(t.startsWith("docusaurus-data-")){var a=t.replace("docusaurus-data-","data-");document.documentElement.setAttribute(a,e)}}catch(t){}}()</script><div id="__docusaurus"><link rel="preload" as="image" href="/solace-agent-mesh/img/logo.png"><link rel="preload" as="image" href="/solace-agent-mesh/img/solace-logo.png"><div role="region" aria-label="Skip to main content"><a class="skipToContent_fXgn" href="#__docusaurus_skipToContent_fallback">Skip to main content</a></div><nav aria-label="Main" class="theme-layout-navbar navbar navbar--fixed-top"><div class="navbar__inner"><div class="theme-layout-navbar-left navbar__items"><button aria-label="Toggle navigation bar" aria-expanded="false" class="navbar__toggle clean-btn" type="button"><svg width="30" height="30" viewBox="0 0 30 30" aria-hidden="true"><path stroke="currentColor" stroke-linecap="round" stroke-miterlimit="10" stroke-width="2" d="M4 7h22M4 15h22M4 23h22"></path></svg></button><a class="navbar__brand" href="/solace-agent-mesh/docs/documentation/getting-started/introduction"><div class="navbar__logo"><img src="/solace-agent-mesh/img/logo.png" alt="Solace Agent Mesh Logo" class="themedComponent_mlkZ themedComponent--light_NVdE"><img src="/solace-agent-mesh/img/logo.png" alt="Solace Agent Mesh Logo" class="themedComponent_mlkZ themedComponent--dark_xIcU"></div><b class="navbar__title text--truncate">Solace Agent Mesh</b></a><a class="navbar__item navbar__link" href="/solace-agent-mesh/docs/documentation/getting-started/introduction">Documentation</a></div><div class="theme-layout-navbar-right navbar__items navbar__items--right"><a href="https://github.com/SolaceLabs/solace-agent-mesh/" target="_blank" rel="noopener noreferrer" class="navbar__item navbar__link">GitHub<svg width="13.5" height="13.5" aria-hidden="true" class="iconExternalLink_nPIU"><use href="#theme-svg-external-link"></use></svg></a><div class="toggle_vylO colorModeToggle_DEke"><button class="clean-btn toggleButton_gllP toggleButtonDisabled_aARS" type="button" disabled="" title="system mode" aria-label="Switch between dark and light mode (currently system mode)"><svg viewBox="0 0 24 24" width="24" height="24" aria-hidden="true" class="toggleIcon_g3eP lightToggleIcon_pyhR"><path fill="currentColor" d="M12,9c1.65,0,3,1.35,3,3s-1.35,3-3,3s-3-1.35-3-3S10.35,9,12,9 M12,7c-2.76,0-5,2.24-5,5s2.24,5,5,5s5-2.24,5-5 S14.76,7,12,7L12,7z M2,13l2,0c0.55,0,1-0.45,1-1s-0.45-1-1-1l-2,0c-0.55,0-1,0.45-1,1S1.45,13,2,13z M20,13l2,0c0.55,0,1-0.45,1-1 s-0.45-1-1-1l-2,0c-0.55,0-1,0.45-1,1S19.45,13,20,13z M11,2v2c0,0.55,0.45,1,1,1s1-0.45,1-1V2c0-0.55-0.45-1-1-1S11,1.45,11,2z M11,20v2c0,0.55,0.45,1,1,1s1-0.45,1-1v-2c0-0.55-0.45-1-1-1C11.45,19,11,19.45,11,20z M5.99,4.58c-0.39-0.39-1.03-0.39-1.41,0 c-0.39,0.39-0.39,1.03,0,1.41l1.06,1.06c0.39,0.39,1.03,0.39,1.41,0s0.39-1.03,0-1.41L5.99,4.58z M18.36,16.95 c-0.39-0.39-1.03-0.39-1.41,0c-0.39,0.39-0.39,1.03,0,1.41l1.06,1.06c0.39,0.39,1.03,0.39,1.41,0c0.39-0.39,0.39-1.03,0-1.41 L18.36,16.95z M19.42,5.99c0.39-0.39,0.39-1.03,0-1.41c-0.39-0.39-1.03-0.39-1.41,0l-1.06,1.06c-0.39,0.39-0.39,1.03,0,1.41 s1.03,0.39,1.41,0L19.42,5.99z M7.05,18.36c0.39-0.39,0.39-1.03,0-1.41c-0.39-0.39-1.03-0.39-1.41,0l-1.06,1.06 c-0.39,0.39-0.39,1.03,0,1.41s1.03,0.39,1.41,0L7.05,18.36z"></path></svg><svg viewBox="0 0 24 24" width="24" height="24" aria-hidden="true" class="toggleIcon_g3eP darkToggleIcon_wfgR"><path fill="currentColor" d="M9.37,5.51C9.19,6.15,9.1,6.82,9.1,7.5c0,4.08,3.32,7.4,7.4,7.4c0.68,0,1.35-0.09,1.99-0.27C17.45,17.19,14.93,19,12,19 c-3.86,0-7-3.14-7-7C5,9.07,6.81,6.55,9.37,5.51z M12,3c-4.97,0-9,4.03-9,9s4.03,9,9,9s9-4.03,9-9c0-0.46-0.04-0.92-0.1-1.36 c-0.98,1.37-2.58,2.26-4.4,2.26c-2.98,0-5.4-2.42-5.4-5.4c0-1.81,0.89-3.42,2.26-4.4C12.92,3.04,12.46,3,12,3L12,3z"></path></svg><svg viewBox="0 0 24 24" width="24" height="24" aria-hidden="true" class="toggleIcon_g3eP systemToggleIcon_QzmC"><path fill="currentColor" d="m12 21c4.971 0 9-4.029 9-9s-4.029-9-9-9-9 4.029-9 9 4.029 9 9 9zm4.95-13.95c1.313 1.313 2.05 3.093 2.05 4.95s-0.738 3.637-2.05 4.95c-1.313 1.313-3.093 2.05-4.95 2.05v-14c1.857 0 3.637 0.737 4.95 2.05z"></path></svg></button></div><div class="navbarSearchContainer_Bca1"><div class="navbar__search"><span aria-label="expand searchbar" role="button" class="search-icon" tabindex="0"></span><input id="search_input_react" type="search" placeholder="Loading..." aria-label="Search" class="navbar__search-input search-bar" disabled=""></div></div></div></div><div role="presentation" class="navbar-sidebar__backdrop"></div></nav><div id="__docusaurus_skipToContent_fallback" class="theme-layout-main main-wrapper mainWrapper_z2l0"><main class="container margin-vert--xl"><div class="row"><div class="col col--6 col--offset-3"><h1 class="hero__title">Page Not Found</h1><p>We could not find what you were looking for.</p><p>Please contact the owner of the site that linked you to the original URL and let them know their link is broken.</p></div></div></main></div><footer class="theme-layout-footer footer footer--dark"><div class="container container-fluid"><div class="row footer__links"><div class="theme-layout-footer-column col footer__col"><div class="footer__title">Solace Agent Mesh</div><ul class="footer__items clean-list"><li class="footer__item"><a class="footer__link-item" href="/solace-agent-mesh/docs/documentation/getting-started/introduction">Documentation</a></li><li class="footer__item"><a href="https://github.com/SolaceLabs/solace-agent-mesh/" target="_blank" rel="noopener noreferrer" class="footer__link-item">GitHub<svg width="13.5" height="13.5" aria-hidden="true" class="iconExternalLink_nPIU"><use href="#theme-svg-external-link"></use></svg></a></li><li class="footer__item"><a href="https://github.com/SolaceLabs/solace-agent-mesh-core-plugins/" target="_blank" rel="noopener noreferrer" class="footer__link-item">Official Plugins<svg width="13.5" height="13.5" aria-hidden="true" class="iconExternalLink_nPIU"><use href="#theme-svg-external-link"></use></svg></a></li></ul></div><div class="theme-layout-footer-column col footer__col"><div class="footer__title">Company</div><ul class="footer__items clean-list"><li class="footer__item"><a href="https://solace.com/products/" target="_blank" rel="noopener noreferrer" class="footer__link-item">Products<svg width="13.5" height="13.5" aria-hidden="true" class="iconExternalLink_nPIU"><use href="#theme-svg-external-link"></use></svg></a></li><li class="footer__item"><a href="https://solace.com/contact/" target="_blank" rel="noopener noreferrer" class="footer__link-item">Contact<svg width="13.5" height="13.5" aria-hidden="true" class="iconExternalLink_nPIU"><use href="#theme-svg-external-link"></use></svg></a></li><li class="footer__item"><a href="https://solace.com/support/" target="_blank" rel="noopener noreferrer" class="footer__link-item">Support<svg width="13.5" height="13.5" aria-hidden="true" class="iconExternalLink_nPIU"><use href="#theme-svg-external-link"></use></svg></a></li><li class="footer__item"><a href="https://solace.com/legal/" target="_blank" rel="noopener noreferrer" class="footer__link-item">Privacy and Legal<svg width="13.5" height="13.5" aria-hidden="true" class="iconExternalLink_nPIU"><use href="#theme-svg-external-link"></use></svg></a></li></ul></div><div class="theme-layout-footer-column col footer__col"><div class="footer__title">Community</div><ul class="footer__items clean-list"><li class="footer__item"><a href="https://www.linkedin.com/company/solacedotcom/" target="_blank" rel="noopener noreferrer" class="footer__link-item">LinkedIn<svg width="13.5" height="13.5" aria-hidden="true" class="iconExternalLink_nPIU"><use href="#theme-svg-external-link"></use></svg></a></li><li class="footer__item"><a href="https://github.com/SolaceLabs" target="_blank" rel="noopener noreferrer" class="footer__link-item">GitHub<svg width="13.5" height="13.5" aria-hidden="true" class="iconExternalLink_nPIU"><use href="#theme-svg-external-link"></use></svg></a></li><li class="footer__item"><a href="https://www.youtube.com/SolaceSystems" target="_blank" rel="noopener noreferrer" class="footer__link-item">YouTube<svg width="13.5" height="13.5" aria-hidden="true" class="iconExternalLink_nPIU"><use href="#theme-svg-external-link"></use></svg></a></li><li class="footer__item"><a href="https://twitter.com/solacedotcom" target="_blank" rel="noopener noreferrer" class="footer__link-item">X<svg width="13.5" height="13.5" aria-hidden="true" class="iconExternalLink_nPIU"><use href="#theme-svg-external-link"></use></svg></a></li></ul></div></div><div class="footer__bottom text--center"><div class="margin-bottom--sm"><img src="/solace-agent-mesh/img/solace-logo.png" alt="Solace Logo" class="footer__logo themedComponent_mlkZ themedComponent--light_NVdE" width="10%" height="10%"><img src="/solace-agent-mesh/img/solace-logo.png" alt="Solace Logo" class="footer__logo themedComponent_mlkZ themedComponent--dark_xIcU" width="10%" height="10%"></div><div class="footer__copyright">Solace Agent Mesh. Copyright © 2025 Solace. Version: 1.3.0</div></div></div></footer></div>
 </body>
 </html>

solace_agent_mesh/assets/docs/assets/js/6e0db977.39a79ca9.js ADDED Viewed

@@ -0,0 +1 @@

+ "use strict";(self.webpackChunksolace_agenitc_mesh_docs=self.webpackChunksolace_agenitc_mesh_docs||[]).push([[217],{8197:(e,t,n)=>{n.r(t),n.d(t,{assets:()=>l,contentTitle:()=>o,default:()=>h,frontMatter:()=>i,metadata:()=>a,toc:()=>c});const a=JSON.parse('{"id":"documentation/migration-guides/a2a-upgrade-to-0.3.0/a2a-gateway-upgrade-to-0.3.0","title":"SAM Gateway Migration Guide: Upgrading to the A2A SDK","description":"This guide is for developers who have built or are maintaining a custom Solace Agent Mesh (SAM) gateway. A recent architectural update has aligned SAM with the official Agent-to-Agent (A2A) protocol","source":"@site/docs/documentation/migration-guides/a2a-upgrade-to-0.3.0/a2a-gateway-upgrade-to-0.3.0.md","sourceDirName":"documentation/migration-guides/a2a-upgrade-to-0.3.0","slug":"/documentation/migration-guides/a2a-upgrade-to-0.3.0/a2a-gateway-upgrade-to-0.3.0","permalink":"/solace-agent-mesh/docs/documentation/migration-guides/a2a-upgrade-to-0.3.0/a2a-gateway-upgrade-to-0.3.0","draft":false,"unlisted":false,"editUrl":"https://github.com/SolaceLabs/solace-agent-mesh/edit/main/docs/docs/documentation/migration-guides/a2a-upgrade-to-0.3.0/a2a-gateway-upgrade-to-0.3.0.md","tags":[],"version":"current","frontMatter":{},"sidebar":"docSidebar","previous":{"title":"RAG Integration","permalink":"/solace-agent-mesh/docs/documentation/tutorials/rag-integration"},"next":{"title":"A2A Technical Migration Map","permalink":"/solace-agent-mesh/docs/documentation/migration-guides/a2a-upgrade-to-0.3.0/a2a-technical-migration-map"}}');var s=n(4848),r=n(8453);const i={},o="SAM Gateway Migration Guide: Upgrading to the A2A SDK",l={},c=[{value:"1. Why the Change?",id:"1-why-the-change",level:2},{value:"2. Core Conceptual Changes",id:"2-core-conceptual-changes",level:2},{value:"The <code>a2a</code> Helper Layer: The New Best Practice",id:"the-a2a-helper-layer-the-new-best-practice",level:3},{value:"Type System Migration",id:"type-system-migration",level:3},{value:"Accessing Object Properties",id:"accessing-object-properties",level:3},{value:"Changes to <code>BaseGatewayComponent</code>",id:"changes-to-basegatewaycomponent",level:3},{value:"3. Practical Migration Checklist",id:"3-practical-migration-checklist",level:2},{value:"4. Code Examples: Before & After",id:"4-code-examples-before--after",level:2},{value:"Example 1: Translating External Input",id:"example-1-translating-external-input",level:3},{value:"Example 2: Sending a Final Response",id:"example-2-sending-a-final-response",level:3}];function d(e){const t={br:"br",code:"code",h1:"h1",h2:"h2",h3:"h3",header:"header",li:"li",ol:"ol",p:"p",pre:"pre",strong:"strong",table:"table",tbody:"tbody",td:"td",th:"th",thead:"thead",tr:"tr",ul:"ul",...(0,r.R)(),...e.components},{Details:n}=t;return n||function(e,t){throw new Error("Expected "+(t?"component":"object")+" `"+e+"` to be defined: you likely forgot to import, pass, or provide it.")}("Details",!0),(0,s.jsxs)(s.Fragment,{children:[(0,s.jsx)(t.header,{children:(0,s.jsx)(t.h1,{id:"sam-gateway-migration-guide-upgrading-to-the-a2a-sdk",children:"SAM Gateway Migration Guide: Upgrading to the A2A SDK"})}),"\n",(0,s.jsxs)(t.p,{children:["This guide is for developers who have built or are maintaining a custom Solace Agent Mesh (SAM) gateway. A recent architectural update has aligned SAM with the official Agent-to-Agent (A2A) protocol",(0,s.jsx)(t.br,{}),"\n","specification by adopting the ",(0,s.jsx)(t.code,{children:"a2a-sdk"}),". This migration requires some changes to your gateway code to ensure compatibility."]}),"\n",(0,s.jsx)(t.p,{children:"This document provides a high-level overview of the conceptual changes and a practical checklist to guide you through the upgrade process."}),"\n",(0,s.jsx)(t.h2,{id:"1-why-the-change",children:"1. Why the Change?"}),"\n",(0,s.jsxs)(t.p,{children:["The migration from our legacy A2A implementation to the official ",(0,s.jsx)(t.code,{children:"a2a-sdk"})," is a foundational improvement with several key benefits:"]}),"\n",(0,s.jsxs)(t.ul,{children:["\n",(0,s.jsxs)(t.li,{children:[(0,s.jsx)(t.strong,{children:"Protocol Compliance:"})," Ensures your gateway is fully interoperable with any A2A-compliant agent or system."]}),"\n",(0,s.jsxs)(t.li,{children:[(0,s.jsx)(t.strong,{children:"Standardization:"})," Replaces bespoke code with a community-supported standard, reducing technical debt."]}),"\n",(0,s.jsxs)(t.li,{children:[(0,s.jsx)(t.strong,{children:"Improved Maintainability:"})," Insulates your gateway from future A2A specification changes. The ",(0,s.jsx)(t.code,{children:"a2a-sdk"})," will be updated, not your core logic."]}),"\n",(0,s.jsxs)(t.li,{children:[(0,s.jsx)(t.strong,{children:"Future-Proofing:"})," Positions your gateway to easily adopt new features as the A2A protocol evolves."]}),"\n"]}),"\n",(0,s.jsx)(t.h2,{id:"2-core-conceptual-changes",children:"2. Core Conceptual Changes"}),"\n",(0,s.jsx)(t.p,{children:"The upgrade introduces a few key changes in how you interact with A2A objects."}),"\n",(0,s.jsxs)(t.h3,{id:"the-a2a-helper-layer-the-new-best-practice",children:["The ",(0,s.jsx)(t.code,{children:"a2a"})," Helper Layer: The New Best Practice"]}),"\n",(0,s.jsxs)(t.p,{children:["The most significant change is the introduction of a new abstraction layer located at ",(0,s.jsx)(t.code,{children:"solace_agent_mesh.common.a2a"}),"."]}),"\n",(0,s.jsxs)(t.p,{children:["You should ",(0,s.jsxs)(t.strong,{children:["no longer instantiate ",(0,s.jsx)(t.code,{children:"a2a.types"})," models directly"]})," or access their properties by hand. Instead, use the provided helper functions. This layer is designed to simplify development and",(0,s.jsx)(t.br,{}),"\n","protect your code from future SDK changes."]}),"\n",(0,s.jsx)(t.p,{children:(0,s.jsx)(t.strong,{children:"Example:"})}),"\n",(0,s.jsx)(t.pre,{children:(0,s.jsx)(t.code,{className:"language-python",children:'# BEFORE: Direct instantiation and property access \nfrom solace_agent_mesh.common.types import TextPart, Task \nmy_part = TextPart(text="Hello") \ntask_id = my_task.id \n \n# AFTER: Using the a2a helper layer \nfrom solace_agent_mesh.common import a2a \nmy_part = a2a.create_text_part(text="Hello") \ntask_id = a2a.get_task_id(my_task) \n'})}),"\n",(0,s.jsx)(t.h3,{id:"type-system-migration",children:"Type System Migration"}),"\n",(0,s.jsxs)(t.ul,{children:["\n",(0,s.jsxs)(t.li,{children:["The legacy types in ",(0,s.jsx)(t.code,{children:"solace_agent_mesh.common.types"})," (like ",(0,s.jsx)(t.code,{children:"A2APart"}),", ",(0,s.jsx)(t.code,{children:"FileContent"}),") are deprecated."]}),"\n",(0,s.jsxs)(t.li,{children:["All A2A models now come from the ",(0,s.jsx)(t.code,{children:"a2a.types"})," library."]}),"\n",(0,s.jsxs)(t.li,{children:["The type hint for a list of message parts has changed from ",(0,s.jsx)(t.code,{children:"List[A2APart]"})," to ",(0,s.jsx)(t.code,{children:"List[ContentPart]"}),". ",(0,s.jsx)(t.code,{children:"ContentPart"})," is a simple alias for the union of ",(0,s.jsx)(t.code,{children:"TextPart"}),", ",(0,s.jsx)(t.code,{children:"DataPart"}),", and ",(0,s.jsx)(t.code,{children:"FilePart"}),"."]}),"\n"]}),"\n",(0,s.jsx)(t.h3,{id:"accessing-object-properties",children:"Accessing Object Properties"}),"\n",(0,s.jsxs)(t.p,{children:["Field names on many A2A objects have changed. Always use the ",(0,s.jsx)(t.code,{children:"a2a"})," helper functions for safe and future-proof access."]}),"\n",(0,s.jsxs)(t.table,{children:[(0,s.jsx)(t.thead,{children:(0,s.jsxs)(t.tr,{children:[(0,s.jsx)(t.th,{style:{textAlign:"left"},children:"Action"}),(0,s.jsx)(t.th,{style:{textAlign:"left"},children:"Old Pattern"}),(0,s.jsx)(t.th,{style:{textAlign:"left"},children:"New Pattern (Recommended)"})]})}),(0,s.jsxs)(t.tbody,{children:[(0,s.jsxs)(t.tr,{children:[(0,s.jsx)(t.td,{style:{textAlign:"left"},children:"Get Task ID"}),(0,s.jsx)(t.td,{style:{textAlign:"left"},children:(0,s.jsx)(t.code,{children:"task.id"})}),(0,s.jsx)(t.td,{style:{textAlign:"left"},children:(0,s.jsx)(t.code,{children:"a2a.get_task_id(task)"})})]}),(0,s.jsxs)(t.tr,{children:[(0,s.jsx)(t.td,{style:{textAlign:"left"},children:"Get Task Status"}),(0,s.jsx)(t.td,{style:{textAlign:"left"},children:(0,s.jsx)(t.code,{children:"task.status.state"})}),(0,s.jsx)(t.td,{style:{textAlign:"left"},children:(0,s.jsx)(t.code,{children:"a2a.get_task_status(task)"})})]}),(0,s.jsxs)(t.tr,{children:[(0,s.jsx)(t.td,{style:{textAlign:"left"},children:"Get Event's Task ID"}),(0,s.jsx)(t.td,{style:{textAlign:"left"},children:(0,s.jsx)(t.code,{children:"event.id"})}),(0,s.jsx)(t.td,{style:{textAlign:"left"},children:(0,s.jsx)(t.code,{children:"event.task_id"})})]}),(0,s.jsxs)(t.tr,{children:[(0,s.jsx)(t.td,{style:{textAlign:"left"},children:"Get Message Parts"}),(0,s.jsx)(t.td,{style:{textAlign:"left"},children:(0,s.jsx)(t.code,{children:"message.parts"})}),(0,s.jsx)(t.td,{style:{textAlign:"left"},children:(0,s.jsx)(t.code,{children:"a2a.get_parts_from_message(message)"})})]}),(0,s.jsxs)(t.tr,{children:[(0,s.jsx)(t.td,{style:{textAlign:"left"},children:"Get Error Message"}),(0,s.jsx)(t.td,{style:{textAlign:"left"},children:(0,s.jsx)(t.code,{children:"error.message"})}),(0,s.jsx)(t.td,{style:{textAlign:"left"},children:(0,s.jsx)(t.code,{children:"a2a.get_error_message(error)"})})]}),(0,s.jsxs)(t.tr,{children:[(0,s.jsx)(t.td,{style:{textAlign:"left"},children:"Get Error Code"}),(0,s.jsx)(t.td,{style:{textAlign:"left"},children:(0,s.jsx)(t.code,{children:"error.code"})}),(0,s.jsx)(t.td,{style:{textAlign:"left"},children:(0,s.jsx)(t.code,{children:"a2a.get_error_code(error)"})})]}),(0,s.jsxs)(t.tr,{children:[(0,s.jsx)(t.td,{style:{textAlign:"left"},children:"Get Error Data"}),(0,s.jsx)(t.td,{style:{textAlign:"left"},children:(0,s.jsx)(t.code,{children:"error.data"})}),(0,s.jsx)(t.td,{style:{textAlign:"left"},children:(0,s.jsx)(t.code,{children:"a2a.get_error_data(error)"})})]})]})]}),"\n",(0,s.jsxs)(t.h3,{id:"changes-to-basegatewaycomponent",children:["Changes to ",(0,s.jsx)(t.code,{children:"BaseGatewayComponent"})]}),"\n",(0,s.jsxs)(t.p,{children:["The ",(0,s.jsx)(t.code,{children:"BaseGatewayComponent"})," has been significantly improved. It now handles more of the A2A protocol complexity, simplifying gateway implementations."]}),"\n",(0,s.jsxs)(t.ul,{children:["\n",(0,s.jsxs)(t.li,{children:[(0,s.jsx)(t.strong,{children:"Artifact Handling:"})," The base class can now automatically handle artifact URIs, converting them to embedded bytes before sending them to your gateway's ",(0,s.jsx)(t.code,{children:"_send_..."})," methods. This is controlled by",(0,s.jsx)(t.br,{}),"\n","the ",(0,s.jsx)(t.code,{children:"resolve_artifact_uris_in_gateway"})," parameter in the constructor."]}),"\n",(0,s.jsxs)(t.li,{children:[(0,s.jsx)(t.strong,{children:"Message Publishing:"})," The base class now manages the details of preparing and publishing the A2A request message in ",(0,s.jsx)(t.code,{children:"submit_a2a_task"}),"."]}),"\n",(0,s.jsxs)(t.li,{children:[(0,s.jsx)(t.strong,{children:"Asynchronous Model:"})," The underlying threading model has been removed in favor of a more direct ",(0,s.jsx)(t.code,{children:"asyncio"})," implementation, simplifying the component lifecycle."]}),"\n"]}),"\n",(0,s.jsxs)(t.p,{children:["It is highly recommended to review the latest ",(0,s.jsx)(t.code,{children:"BaseGatewayComponent"})," and re-base your custom gateway on it to inherit these benefits and reduce boilerplate code."]}),"\n",(0,s.jsx)(t.h2,{id:"3-practical-migration-checklist",children:"3. Practical Migration Checklist"}),"\n",(0,s.jsx)(t.p,{children:"Follow these steps to update your gateway code."}),"\n",(0,s.jsxs)(t.ol,{children:["\n",(0,s.jsxs)(t.li,{children:["\n",(0,s.jsx)(t.p,{children:(0,s.jsx)(t.strong,{children:"Update Imports:"})}),"\n",(0,s.jsxs)(t.ul,{children:["\n",(0,s.jsxs)(t.li,{children:["Remove imports from ",(0,s.jsx)(t.code,{children:"solace_agent_mesh.common.types"}),"."]}),"\n",(0,s.jsxs)(t.li,{children:["Add imports from ",(0,s.jsx)(t.code,{children:"a2a.types"})," for specific models if needed."]}),"\n",(0,s.jsxs)(t.li,{children:["Add ",(0,s.jsx)(t.code,{children:"from solace_agent_mesh.common import a2a"}),"."]}),"\n",(0,s.jsxs)(t.li,{children:["Add ",(0,s.jsx)(t.code,{children:"from solace_agent_mesh.common.a2a import ContentPart"}),"."]}),"\n"]}),"\n"]}),"\n",(0,s.jsxs)(t.li,{children:["\n",(0,s.jsx)(t.p,{children:(0,s.jsx)(t.strong,{children:"Update Type Hints:"})}),"\n",(0,s.jsxs)(t.ul,{children:["\n",(0,s.jsxs)(t.li,{children:["Find all instances of ",(0,s.jsx)(t.code,{children:"List[A2APart]"})," and change them to ",(0,s.jsx)(t.code,{children:"List[ContentPart]"}),"."]}),"\n"]}),"\n"]}),"\n",(0,s.jsxs)(t.li,{children:["\n",(0,s.jsx)(t.p,{children:(0,s.jsx)(t.strong,{children:"Refactor Object Creation:"})}),"\n",(0,s.jsxs)(t.ul,{children:["\n",(0,s.jsxs)(t.li,{children:["Replace direct model instantiation like ",(0,s.jsx)(t.code,{children:"TextPart(...)"})," or ",(0,s.jsx)(t.code,{children:"FilePart(...)"})," with their corresponding helper functions: ",(0,s.jsx)(t.code,{children:"a2a.create_text_part(...)"}),", ",(0,s.jsx)(t.code,{children:"a2a.create_file_part_from_uri(...)"}),", etc."]}),"\n"]}),"\n"]}),"\n",(0,s.jsxs)(t.li,{children:["\n",(0,s.jsx)(t.p,{children:(0,s.jsx)(t.strong,{children:"Refactor Property Access:"})}),"\n",(0,s.jsxs)(t.ul,{children:["\n",(0,s.jsxs)(t.li,{children:["Replace direct property access (",(0,s.jsx)(t.code,{children:"task.id"}),", ",(0,s.jsx)(t.code,{children:"error.message"}),") with calls to the ",(0,s.jsx)(t.code,{children:"a2a"})," helper functions (",(0,s.jsx)(t.code,{children:"a2a.get_task_id(task)"}),", ",(0,s.jsx)(t.code,{children:"a2a.get_error_message(error)"}),")."]}),"\n"]}),"\n"]}),"\n",(0,s.jsxs)(t.li,{children:["\n",(0,s.jsx)(t.p,{children:(0,s.jsx)(t.strong,{children:"Review Base Class Integration:"})}),"\n",(0,s.jsxs)(t.ul,{children:["\n",(0,s.jsxs)(t.li,{children:["If you have a custom gateway, compare it against the latest ",(0,s.jsx)(t.code,{children:"BaseGatewayComponent"}),". Consider refactoring to delegate more responsibility (like artifact handling and message submission) to the",(0,s.jsx)(t.br,{}),"\n","base class."]}),"\n"]}),"\n"]}),"\n",(0,s.jsxs)(t.li,{children:["\n",(0,s.jsx)(t.p,{children:(0,s.jsx)(t.strong,{children:"Test Thoroughly:"})}),"\n",(0,s.jsxs)(t.ul,{children:["\n",(0,s.jsx)(t.li,{children:"Once refactored, run your integration tests to ensure the gateway correctly translates inputs and processes outputs in the new format."}),"\n"]}),"\n"]}),"\n"]}),"\n",(0,s.jsx)(t.h2,{id:"4-code-examples-before--after",children:"4. Code Examples: Before & After"}),"\n",(0,s.jsx)(t.p,{children:"Here are some common patterns you will encounter during the migration."}),"\n",(0,s.jsx)(t.h3,{id:"example-1-translating-external-input",children:"Example 1: Translating External Input"}),"\n",(0,s.jsx)(t.p,{children:"This example shows how to create A2A parts from an external event."}),"\n",(0,s.jsxs)(n,{children:[(0,s.jsx)("summary",{children:(0,s.jsx)("strong",{children:"_translate_external_input"})}),(0,s.jsx)(t.p,{children:(0,s.jsx)(t.strong,{children:"Before:"})}),(0,s.jsx)(t.pre,{children:(0,s.jsx)(t.code,{className:"language-python",children:'from solace_agent_mesh.common.types import Part as A2APart, TextPart, FilePart, FileContent \n \nasync def _translate_external_input(self, external_event: Any) -> Tuple[str, List[A2APart], Dict[str, Any]]: \n # ... \n a2a_parts: List[A2APart] = [] \n \n # Create a file part with a URI \n uri = "artifact://..." \n a2a_parts.append( \n FilePart( \n file=FileContent(name="report.pdf", uri=uri) \n ) \n ) \n \n # Create a text part \n prompt = "Summarize the attached file." \n a2a_parts.append(TextPart(text=prompt)) \n \n return "summary_agent", a2a_parts, {} \n'})}),(0,s.jsx)(t.p,{children:(0,s.jsx)(t.strong,{children:"After:"})}),(0,s.jsx)(t.pre,{children:(0,s.jsx)(t.code,{className:"language-python",children:'from solace_agent_mesh.common import a2a \nfrom solace_agent_mesh.common.a2a import ContentPart \n \nasync def _translate_external_input(self, external_event: Any) -> Tuple[str, List[ContentPart], Dict[str, Any]]: \n # ... \n a2a_parts: List[ContentPart] = [] \n \n # Create a file part with a URI using the helper \n uri = "artifact://..." \n file_part = a2a.create_file_part_from_uri(uri=uri, name="report.pdf") \n a2a_parts.append(file_part) \n \n # Create a text part using the helper \n prompt = "Summarize the attached file." \n text_part = a2a.create_text_part(text=prompt) \n a2a_parts.append(text_part) \n \n return "summary_agent", a2a_parts, {} \n'})})]}),"\n",(0,s.jsx)(t.h3,{id:"example-2-sending-a-final-response",children:"Example 2: Sending a Final Response"}),"\n",(0,s.jsxs)(t.p,{children:["This example shows how to process a final ",(0,s.jsx)(t.code,{children:"Task"})," object."]}),"\n",(0,s.jsxs)(n,{children:[(0,s.jsx)("summary",{children:(0,s.jsx)("strong",{children:"_send_final_response_to_external"})}),(0,s.jsx)(t.p,{children:(0,s.jsx)(t.strong,{children:"Before:"})}),(0,s.jsx)(t.pre,{children:(0,s.jsx)(t.code,{className:"language-python",children:'from solace_agent_mesh.common.types import Task, TaskState, TextPart \n \nasync def _send_final_response_to_external(self, context: Dict, task_data: Task) -> None: \n task_id = task_data.id \n final_status_text = ":checkered_flag: Task complete." \n \n if task_data.status.state == TaskState.FAILED: \n error_message_text = "" \n if task_data.status.message and task_data.status.message.parts: \n for part in task_data.status.message.parts: \n if isinstance(part, TextPart): \n error_message_text = part.text \n break \n final_status_text = f":x: Error: Task failed. {error_message_text}".strip() \n \n # ... use final_status_text and task_id \n'})}),(0,s.jsx)(t.p,{children:(0,s.jsx)(t.strong,{children:"After:"})}),(0,s.jsx)(t.pre,{children:(0,s.jsx)(t.code,{className:"language-python",children:'from solace_agent_mesh.common import a2a \nfrom a2a.types import Task, TaskState \n \nasync def _send_final_response_to_external(self, context: Dict, task_data: Task) -> None: \n # Use helpers to safely access properties \n task_id = a2a.get_task_id(task_data) \n task_status = a2a.get_task_status(task_data) \n \n final_status_text = ":checkered_flag: Task complete." \n \n if task_status == TaskState.failed: \n error_message_text = "" \n if task_data.status and task_data.status.message: \n # Use helper to extract all text from the message \n error_message_text = a2a.get_text_from_message(task_data.status.message) \n final_status_text = f":x: Error: Task failed. {error_message_text}".strip() \n \n # ... use final_status_text and task_id \n'})})]})]})}function h(e={}){const{wrapper:t}={...(0,r.R)(),...e.components};return t?(0,s.jsx)(t,{...e,children:(0,s.jsx)(d,{...e})}):d(e)}},8453:(e,t,n)=>{n.d(t,{R:()=>i,x:()=>o});var a=n(6540);const s={},r=a.createContext(s);function i(e){const t=a.useContext(r);return a.useMemo((function(){return"function"==typeof e?e(t):{...t,...e}}),[t,e])}function o(e){let t;return t=e.disableParentContext?"function"==typeof e.components?e.components(s):e.components||s:i(e.components),a.createElement(r.Provider,{value:t},e.children)}}}]);

solace_agent_mesh/assets/docs/assets/js/{75384d09.ccd480c4.js → 75384d09.bf78fbdb.js} RENAMED Viewed

	@@ -1 +1 @@
1	- "use strict";(self.webpackChunksolace_agenitc_mesh_docs=self.webpackChunksolace_agenitc_mesh_docs\|\|[]).push([[8055],{5343:(e,n,t)=>{t.r(n),t.d(n,{assets:()=>l,contentTitle:()=>s,default:()=>h,frontMatter:()=>o,metadata:()=>i,toc:()=>c});const i=JSON.parse('{"id":"documentation/tutorials/rag-integration","title":"RAG Integration","description":"This tutorial guides you through setting up and configuring Solace Agent Mesh (SAM) Retrieval Augmented Generation (RAG) plugin. The RAG plugin enables your agents to answer questions by retrieving information from a knowledge base of your documents.","source":"@site/docs/documentation/tutorials/rag-integration.md","sourceDirName":"documentation/tutorials","slug":"/documentation/tutorials/rag-integration","permalink":"/solace-agent-mesh/docs/documentation/tutorials/rag-integration","draft":false,"unlisted":false,"editUrl":"https://github.com/SolaceLabs/solace-agent-mesh/edit/main/docs/docs/documentation/tutorials/rag-integration.md","tags":[],"version":"current","sidebarPosition":70,"frontMatter":{"title":"RAG Integration","sidebar_position":70,"toc_max_heading_level":4},"sidebar":"docSidebar","previous":{"title":"Slack Integration","permalink":"/solace-agent-mesh/docs/documentation/tutorials/slack-integration"}}');var a=t(4848),r=t(8453);const o={title:"RAG Integration",sidebar_position:70,toc_max_heading_level:4},s="RAG Integration",l={},c=[{value:"What is SAM RAG?",id:"what-is-sam-rag",level:2},{value:"Prerequisites",id:"prerequisites",level:2},{value:"Adding the RAG Plugin",id:"adding-the-rag-plugin",level:2},{value:"Configuring the RAG Agent",id:"configuring-the-rag-agent",level:2},{value:"Shared Configuration",id:"shared-configuration",level:3},{value:"RAG Pipeline Configuration",id:"rag-pipeline-configuration",level:3},{value:"1. Scanner Configuration",id:"1-scanner-configuration",level:4},{value:"2. Preprocessor Configuration",id:"2-preprocessor-configuration",level:4},{value:"3. Splitter Configuration",id:"3-splitter-configuration",level:4},{value:"4. Embedding Configuration",id:"4-embedding-configuration",level:4},{value:"5. Vector Database Configuration",id:"5-vector-database-configuration",level:4},{value:"6. LLM Configuration",id:"6-llm-configuration",level:4},{value:"7. Retrieval Configuration",id:"7-retrieval-configuration",level:4},{value:"Environment Variables",id:"environment-variables",level:3},{value:"Running the RAG Agent",id:"running-the-rag-agent",level:2},{value:"Testing the RAG Agent",id:"testing-the-rag-agent",level:2},{value:"Ingesting Documents",id:"ingesting-documents",level:3},{value:"Option 1: Automatic Scanning (Batch Ingestion)",id:"option-1-automatic-scanning-batch-ingestion",level:4},{value:"Option 2: Manual Upload via Gateway",id:"option-2-manual-upload-via-gateway",level:4},{value:"Querying the Knowledge Base",id:"querying-the-knowledge-base",level:3},{value:"Troubleshooting",id:"troubleshooting",level:2}];function d(e){const n={a:"a",blockquote:"blockquote",code:"code",h1:"h1",h2:"h2",h3:"h3",h4:"h4",header:"header",li:"li",ol:"ol",p:"p",pre:"pre",strong:"strong",ul:"ul",...(0,r.R)(),...e.components};return(0,a.jsxs)(a.Fragment,{children:[(0,a.jsx)(n.header,{children:(0,a.jsx)(n.h1,{id:"rag-integration",children:"RAG Integration"})}),"\n",(0,a.jsx)(n.p,{children:"This tutorial guides you through setting up and configuring Solace Agent Mesh (SAM) Retrieval Augmented Generation (RAG) plugin. The RAG plugin enables your agents to answer questions by retrieving information from a knowledge base of your documents."}),"\n",(0,a.jsx)(n.h2,{id:"what-is-sam-rag",children:"What is SAM RAG?"}),"\n",(0,a.jsx)(n.p,{children:"The SAM RAG plugin enhances your agents with the ability to perform retrieval-augmented generation. This means the agent can:"}),"\n",(0,a.jsxs)(n.ul,{children:["\n",(0,a.jsxs)(n.li,{children:[(0,a.jsx)(n.strong,{children:"Scan"})," documents from various sources (local filesystem, Google Drive, etc.)."]}),"\n",(0,a.jsxs)(n.li,{children:[(0,a.jsx)(n.strong,{children:"Preprocess"})," and ",(0,a.jsx)(n.strong,{children:"split"})," the text into manageable chunks."]}),"\n",(0,a.jsxs)(n.li,{children:[(0,a.jsx)(n.strong,{children:"Embed"})," these chunks into vectors and store them in a vector database."]}),"\n",(0,a.jsxs)(n.li,{children:[(0,a.jsx)(n.strong,{children:"Retrieve"})," relevant chunks of text based on a user's query."]}),"\n",(0,a.jsxs)(n.li,{children:[(0,a.jsx)(n.strong,{children:"Generate"})," an answer using a large language model (LLM) augmented with the retrieved information."]}),"\n"]}),"\n",(0,a.jsx)(n.p,{children:"This allows you to build agents that can answer questions about your own private data."}),"\n",(0,a.jsx)(n.h2,{id:"prerequisites",children:"Prerequisites"}),"\n",(0,a.jsx)(n.p,{children:"Before you begin, ensure you have:"}),"\n",(0,a.jsxs)(n.ul,{children:["\n",(0,a.jsxs)(n.li,{children:[(0,a.jsx)(n.a,{href:"/solace-agent-mesh/docs/documentation/getting-started/installation",children:"Installed Solace Agent Mesh and the SAM CLI"}),"."]}),"\n",(0,a.jsxs)(n.li,{children:[(0,a.jsx)(n.a,{href:"/solace-agent-mesh/docs/documentation/getting-started/quick-start",children:"Created a new Solace Agent Mesh project"}),"."]}),"\n",(0,a.jsx)(n.li,{children:"Access to a vector database (for example, Qdrant, Chroma, and Pinecone)."}),"\n",(0,a.jsx)(n.li,{children:"Access to an LLM for generation and an embedding model."}),"\n",(0,a.jsx)(n.li,{children:"A directory with some documents for the agent to ingest."}),"\n"]}),"\n",(0,a.jsx)(n.h2,{id:"adding-the-rag-plugin",children:"Adding the RAG Plugin"}),"\n",(0,a.jsx)(n.p,{children:"To add the RAG plugin to your SAM project, run the following command:"}),"\n",(0,a.jsx)(n.pre,{children:(0,a.jsx)(n.code,{className:"language-sh",children:"sam plugin add my-rag-agent --plugin sam-rag\n"})}),"\n",(0,a.jsxs)(n.p,{children:["Replace ",(0,a.jsx)(n.code,{children:"my-rag-agent"})," with your preferred agent name. This command:"]}),"\n",(0,a.jsxs)(n.ul,{children:["\n",(0,a.jsxs)(n.li,{children:["Installs the ",(0,a.jsx)(n.code,{children:"sam-rag"})," plugin."]}),"\n",(0,a.jsxs)(n.li,{children:["Creates a new agent configuration file at ",(0,a.jsx)(n.code,{children:"configs/agents/my-rag-agent.yaml"}),"."]}),"\n"]}),"\n",(0,a.jsx)(n.h2,{id:"configuring-the-rag-agent",children:"Configuring the RAG Agent"}),"\n",(0,a.jsxs)(n.p,{children:["The RAG agent requires a detailed configuration. Open ",(0,a.jsx)(n.code,{children:"configs/agents/my-rag-agent.yaml"})," to configure the following sections:"]}),"\n",(0,a.jsx)(n.h3,{id:"shared-configuration",children:"Shared Configuration"}),"\n",(0,a.jsxs)(n.p,{children:["Like other agents, the RAG agent needs a connection to the Solace broker and a configured LLM. This is typically done in a ",(0,a.jsx)(n.code,{children:"shared_config.yaml"})," file."]}),"\n",(0,a.jsx)(n.pre,{children:(0,a.jsx)(n.code,{className:"language-yaml",children:"# configs/shared_config.yaml\nshared_config:\n - broker_connection: &broker_connection\n dev_mode: ${SOLACE_DEV_MODE, false}\n broker_url: ${SOLACE_BROKER_URL, ws://localhost:8080}\n broker_username: ${SOLACE_BROKER_USERNAME, default}\n broker_password: ${SOLACE_BROKER_PASSWORD, default}\n broker_vpn: ${SOLACE_BROKER_VPN, default}\n temporary_queue: ${USE_TEMPORARY_QUEUES, true}\n\n - models:\n general: &general_model\n model: ${LLM_SERVICE_GENERAL_MODEL_NAME}\n api_base: ${LLM_SERVICE_ENDPOINT}\n api_key: ${LLM_SERVICE_API_KEY}\n"})}),"\n",(0,a.jsx)(n.h3,{id:"rag-pipeline-configuration",children:"RAG Pipeline Configuration"}),"\n",(0,a.jsxs)(n.p,{children:["The RAG pipeline has several stages, each with its own configuration block within the ",(0,a.jsx)(n.code,{children:"app_config"})," section of your ",(0,a.jsx)(n.code,{children:"my-rag-agent.yaml"})," file."]}),"\n",(0,a.jsx)(n.h4,{id:"1-scanner-configuration",children:"1. Scanner Configuration"}),"\n",(0,a.jsx)(n.p,{children:"The scanner discovers documents to be ingested. You can configure it to scan a local filesystem or cloud sources."}),"\n",(0,a.jsx)(n.p,{children:(0,a.jsx)(n.strong,{children:"Local Filesystem Example:"})}),"\n",(0,a.jsx)(n.pre,{children:(0,a.jsx)(n.code,{className:"language-yaml",children:'scanner:\n batch: true\n use_memory_storage: true\n source:\n type: filesystem\n directories:\n - "/path/to/your/documents" # Important: Replace with your actual document directory path\n filters:\n file_formats:\n - ".txt"\n - ".pdf"\n - ".md"\n'})}),"\n",(0,a.jsxs)(n.p,{children:[(0,a.jsx)(n.strong,{children:"Multi-Cloud Source Example:"}),"\nYou can also configure multiple sources, including Google Drive, OneDrive, and S3."]}),"\n",(0,a.jsx)(n.pre,{children:(0,a.jsx)(n.code,{className:"language-yaml",children:'scanner:\n batch: true\n use_memory_storage: true\n sources:\n - type: filesystem\n directories: ["${LOCAL_DOCUMENTS_PATH}"]\n - type: google_drive\n credentials_path: "${GOOGLE_DRIVE_CREDENTIALS_PATH}"\n folders:\n - folder_id: "${GOOGLE_DRIVE_FOLDER_ID_1}"\n name: "Documents"\n recursive: true\n'})}),"\n",(0,a.jsx)(n.h4,{id:"2-preprocessor-configuration",children:"2. Preprocessor Configuration"}),"\n",(0,a.jsx)(n.p,{children:"The preprocessor cleans the text extracted from documents."}),"\n",(0,a.jsx)(n.pre,{children:(0,a.jsx)(n.code,{className:"language-yaml",children:"preprocessor:\n default_preprocessor:\n type: enhanced\n params:\n lowercase: true\n normalize_whitespace: true\n remove_urls: true\n preprocessors:\n pdf: \n type: document\n params:\n lowercase: true\n normalize_whitespace: true\n remove_non_ascii: true\n remove_urls: true\n"})}),"\n",(0,a.jsx)(n.h4,{id:"3-splitter-configuration",children:"3. Splitter Configuration"}),"\n",(0,a.jsx)(n.p,{children:"The splitter breaks down large documents into smaller chunks. Different splitters are available for different file types."}),"\n",(0,a.jsx)(n.pre,{children:(0,a.jsx)(n.code,{className:"language-yaml",children:"splitter:\n default_splitter:\n type: recursive_character\n params:\n chunk_size: 2048\n chunk_overlap: 400\n splitters:\n markdown:\n type: markdown\n params:\n chunk_size: 2048\n chunk_overlap: 400\n pdf:\n type: token\n params:\n chunk_size: 500\n chunk_overlap: 100\n"})}),"\n",(0,a.jsx)(n.h4,{id:"4-embedding-configuration",children:"4. Embedding Configuration"}),"\n",(0,a.jsx)(n.p,{children:"This section defines the model used to create vector embeddings from the text chunks."}),"\n",(0,a.jsx)(n.pre,{children:(0,a.jsx)(n.code,{className:"language-yaml",children:'embedding:\n embedder_type: "openai"\n embedder_params:\n model: "${OPENAI_EMBEDDING_MODEL}"\n api_key: "${OPENAI_API_KEY}"\n api_base: "${OPENAI_API_ENDPOINT}"\n normalize_embeddings: true\n'})}),"\n",(0,a.jsx)(n.h4,{id:"5-vector-database-configuration",children:"5. Vector Database Configuration"}),"\n",(0,a.jsx)(n.p,{children:"Configure the connection to your vector database where the embeddings are stored."}),"\n",(0,a.jsx)(n.p,{children:(0,a.jsx)(n.strong,{children:"Qdrant Example:"})}),"\n",(0,a.jsx)(n.pre,{children:(0,a.jsx)(n.code,{className:"language-yaml",children:'vector_db:\n db_type: "qdrant"\n db_params:\n url: "${QDRANT_URL}"\n api_key: "${QDRANT_API_KEY}"\n collection_name: "${QDRANT_COLLECTION}"\n embedding_dimension: ${QDRANT_EMBEDDING_DIMENSION}\n'})}),"\n",(0,a.jsx)(n.p,{children:(0,a.jsx)(n.strong,{children:"Chroma Example:"})}),"\n",(0,a.jsx)(n.pre,{children:(0,a.jsx)(n.code,{className:"language-yaml",children:'vector_db:\n db_type: "chroma"\n db_params:\n host: "${CHROMA_HOST}"\n port: "${CHROMA_PORT}"\n collection_name: "${CHROMA_COLLECTION}"\n'})}),"\n",(0,a.jsx)(n.h4,{id:"6-llm-configuration",children:"6. LLM Configuration"}),"\n",(0,a.jsx)(n.p,{children:"Configure the LLM that is used to generate answers based on the retrieved context."}),"\n",(0,a.jsx)(n.pre,{children:(0,a.jsx)(n.code,{className:"language-yaml",children:'llm:\n load_balancer:\n - model_name: "gpt-4o"\n litellm_params:\n model: "openai/${OPENAI_MODEL_NAME}"\n api_key: "${OPENAI_API_KEY}"\n api_base: "${OPENAI_API_ENDPOINT}"\n'})}),"\n",(0,a.jsx)(n.h4,{id:"7-retrieval-configuration",children:"7. Retrieval Configuration"}),"\n",(0,a.jsx)(n.p,{children:"This defines how many document chunks are retrieved to answer a query."}),"\n",(0,a.jsx)(n.pre,{children:(0,a.jsx)(n.code,{className:"language-yaml",children:"retrieval:\n top_k: 7\n"})}),"\n",(0,a.jsx)(n.h3,{id:"environment-variables",children:"Environment Variables"}),"\n",(0,a.jsxs)(n.p,{children:["The RAG agent relies heavily on environment variables. Here are some of the most important ones you'll need to set in your ",(0,a.jsx)(n.code,{children:".env"})," file:"]}),"\n",(0,a.jsx)(n.pre,{children:(0,a.jsx)(n.code,{className:"language-bash",children:'# Solace Connection\nSOLACE_BROKER_URL=ws://localhost:8080\nSOLACE_BROKER_VPN=default\nSOLACE_BROKER_USERNAME=default\nSOLACE_BROKER_PASSWORD=default\nNAMESPACE=my-org/dev\n\n# LLM and Embedding Models\nOPENAI_API_KEY="your-openai-api-key"\nOPENAI_API_ENDPOINT="your-openai-api-endpoint"\nOPENAI_MODEL_NAME="model name. E.g., gpt-4o"\nOPENAI_EMBEDDING_MODEL="embedding model name. E.g., text-embedding-3-small"\n\n# Vector Database (Qdrant example)\nQDRANT_URL="Qdrant url"\nQDRANT_API_KEY="API key"\nQDRANT_COLLECTION="my-rag-collection"\nQDRANT_EMBEDDING_DIMENSION=1536 # Depends on your embedding model\n\n# Scanner\nLOCAL_DOCUMENTS_PATH="./my_documents" # Relative path to your documents folder\n'})}),"\n",(0,a.jsxs)(n.p,{children:["Create a directory named ",(0,a.jsx)(n.code,{children:"my_documents"})," in your project root and place some text or markdown files inside it."]}),"\n",(0,a.jsx)(n.h2,{id:"running-the-rag-agent",children:"Running the RAG Agent"}),"\n",(0,a.jsx)(n.p,{children:"Once you have configured your agent and set the environment variables, you can run it:"}),"\n",(0,a.jsx)(n.pre,{children:(0,a.jsx)(n.code,{className:"language-sh",children:"sam run configs/agents/my-rag-agent.yaml\n"})}),"\n",(0,a.jsx)(n.p,{children:"When the agent starts, it begins scanning the documents in the configured source, processing and ingesting them into your vector database. This process may take some time, depending on the number and size of your documents."}),"\n",(0,a.jsx)(n.h2,{id:"testing-the-rag-agent",children:"Testing the RAG Agent"}),"\n",(0,a.jsx)(n.p,{children:"Once your agent is running, you can test its retrieval capabilities and ingest new documents."}),"\n",(0,a.jsx)(n.h3,{id:"ingesting-documents",children:"Ingesting Documents"}),"\n",(0,a.jsx)(n.p,{children:"There are two primary ways to ingest documents into your RAG agent's knowledge base:"}),"\n",(0,a.jsx)(n.h4,{id:"option-1-automatic-scanning-batch-ingestion",children:"Option 1: Automatic Scanning (Batch Ingestion)"}),"\n",(0,a.jsxs)(n.p,{children:["This method uses the configured ",(0,a.jsx)(n.code,{children:"scanner"})," component. The agent automatically ingests documents from the directories specified in your configuration upon startup."]}),"\n",(0,a.jsx)(n.p,{children:(0,a.jsx)(n.strong,{children:"Step 1: Create a Document"})}),"\n",(0,a.jsxs)(n.p,{children:["First, create a simple text file named ",(0,a.jsx)(n.code,{children:"sam_features.txt"})," and add some content to it. For example:"]}),"\n",(0,a.jsx)(n.pre,{children:(0,a.jsx)(n.code,{className:"language-text",children:"Solace Agent Mesh (SAM) is a powerful framework for building AI agents.\nKey features of SAM include:\n- A flexible plugin architecture.\n- Integration with various LLMs and vector databases.\n- Scalable gateways for different communication protocols.\n- An event-driven design based on Solace PubSub+.\n"})}),"\n",(0,a.jsx)(n.p,{children:(0,a.jsx)(n.strong,{children:"Step 2: Place the Document in the Scanned Directory"})}),"\n",(0,a.jsxs)(n.p,{children:['In the "Environment Variables" section, we configured ',(0,a.jsx)(n.code,{children:"LOCAL_DOCUMENTS_PATH"})," to point to a directory (e.g., ",(0,a.jsx)(n.code,{children:"./my_documents"}),")."]}),"\n",(0,a.jsxs)(n.p,{children:["Create this directory in your project's root folder if you haven't already, and move your ",(0,a.jsx)(n.code,{children:"sam_features.txt"})," file into it."]}),"\n",(0,a.jsx)(n.pre,{children:(0,a.jsx)(n.code,{className:"language-sh",children:"mkdir -p my_documents\nmv sam_features.txt my_documents/\n"})}),"\n",(0,a.jsx)(n.p,{children:(0,a.jsx)(n.strong,{children:"Step 3: Run the Agent to Trigger Ingestion"})}),"\n",(0,a.jsx)(n.p,{children:"If your agent is already running, you'll need to restart it to trigger the batch scan. If it's not running, start it now:"}),"\n",(0,a.jsx)(n.pre,{children:(0,a.jsx)(n.code,{className:"language-sh",children:"sam run configs/agents/my-rag-agent.yaml\n"})}),"\n",(0,a.jsx)(n.p,{children:"You will see logs indicating that the file is being processed. Once the agent is running and the initial scan is complete, the document is successfully ingested and ready for retrieval."}),"\n",(0,a.jsx)(n.h4,{id:"option-2-manual-upload-via-gateway",children:"Option 2: Manual Upload via Gateway"}),"\n",(0,a.jsx)(n.p,{children:"You can also ingest documents dynamically by uploading them directly through a gateway, like the Web UI. This is useful for adding single documents without restarting the agent. The RAG agent exposes a tool for this purpose."}),"\n",(0,a.jsx)(n.p,{children:(0,a.jsx)(n.strong,{children:"Step 1: Start the RAG Agent and Web UI"})}),"\n",(0,a.jsx)(n.p,{children:"Ensure both your RAG agent and the Web UI gateway are running."}),"\n",(0,a.jsx)(n.p,{children:(0,a.jsx)(n.strong,{children:"Step 2: Upload a Document in the Web UI"})}),"\n",(0,a.jsxs)(n.ol,{children:["\n",(0,a.jsxs)(n.li,{children:["Open the Web UI (usually at ",(0,a.jsx)(n.a,{href:"http://localhost:8000",children:"http://localhost:8000"}),", or check your gateway configuration for the correct URL) and start a chat with your RAG agent."]}),"\n",(0,a.jsx)(n.li,{children:"Use the file attachment button to select a document from your local machine."}),"\n",(0,a.jsxs)(n.li,{children:["Send a prompt along with the file, instructing the agent to ingest it. For example:","\n",(0,a.jsxs)(n.blockquote,{children:["\n",(0,a.jsx)(n.p,{children:'"Please ingest the attached document into your knowledge base."'}),"\n"]}),"\n"]}),"\n"]}),"\n",(0,a.jsxs)(n.p,{children:["The RAG agent uses its ",(0,a.jsx)(n.code,{children:"built-in"})," ingest_document tool to process the file you uploaded. The file goes through the same preprocessing, splitting, and embedding pipeline as the documents from the automatic scan."]}),"\n",(0,a.jsx)(n.p,{children:(0,a.jsx)(n.strong,{children:"Step 3: Confirm Ingestion"})}),"\n",(0,a.jsx)(n.p,{children:"After the agent confirms that the document has been ingested, you can immediately ask questions about its content."}),"\n",(0,a.jsx)(n.h3,{id:"querying-the-knowledge-base",children:"Querying the Knowledge Base"}),"\n",(0,a.jsx)(n.p,{children:"You can interact with your RAG agent through any gateway, such as the Web UI gateway."}),"\n",(0,a.jsxs)(n.ol,{children:["\n",(0,a.jsx)(n.li,{children:"Make sure you have a Web UI gateway running (or add one to your project)."}),"\n",(0,a.jsxs)(n.li,{children:["Open the Web UI (usually at ",(0,a.jsx)(n.code,{children:"http://localhost:8000"}),")."]}),"\n",(0,a.jsxs)(n.li,{children:["Start a conversation with ",(0,a.jsx)(n.code,{children:"my-rag-agent"}),"."]}),"\n",(0,a.jsx)(n.li,{children:"Ask a question related to the content of the documents you provided during the initial scan."}),"\n"]}),"\n",(0,a.jsx)(n.p,{children:"For example, if you have a document about product features, you could ask:"}),"\n",(0,a.jsxs)(n.blockquote,{children:["\n",(0,a.jsx)(n.p,{children:'"What are the key features of Product X?"'}),"\n"]}),"\n",(0,a.jsx)(n.p,{children:"The agent searches its knowledge base, finds the relevant information, and generates an answer based on the content of your documents."}),"\n",(0,a.jsx)(n.h2,{id:"troubleshooting",children:"Troubleshooting"}),"\n",(0,a.jsxs)(n.ul,{children:["\n",(0,a.jsxs)(n.li,{children:[(0,a.jsx)(n.strong,{children:"Connection Errors"}),": Double-check all your URLs, API keys, and credentials for your LLM and vector database."]}),"\n",(0,a.jsxs)(n.li,{children:[(0,a.jsx)(n.strong,{children:"Ingestion Issues"}),": Check the agent logs for errors during the scanning, preprocessing, or embedding stages. Ensure the file formats are supported and the paths are correct."]}),"\n",(0,a.jsxs)(n.li,{children:[(0,a.jsx)(n.strong,{children:"No Answers"}),": If the agent can't answer, it might be because the information is not in the documents, or the ",(0,a.jsx)(n.code,{children:"top_k"})," retrieval setting is too low."]}),"\n"]})]})}function h(e={}){const{wrapper:n}={...(0,r.R)(),...e.components};return n?(0,a.jsx)(n,{...e,children:(0,a.jsx)(d,{...e})}):d(e)}},8453:(e,n,t)=>{t.d(n,{R:()=>o,x:()=>s});var i=t(6540);const a={},r=i.createContext(a);function o(e){const n=i.useContext(r);return i.useMemo((function(){return"function"==typeof e?e(n):{...n,...e}}),[n,e])}function s(e){let n;return n=e.disableParentContext?"function"==typeof e.components?e.components(a):e.components\|\|a:o(e.components),i.createElement(r.Provider,{value:n},e.children)}}}]);
1	+ "use strict";(self.webpackChunksolace_agenitc_mesh_docs=self.webpackChunksolace_agenitc_mesh_docs\|\|[]).push([[8055],{5343:(e,n,t)=>{t.r(n),t.d(n,{assets:()=>l,contentTitle:()=>s,default:()=>h,frontMatter:()=>o,metadata:()=>i,toc:()=>c});const i=JSON.parse('{"id":"documentation/tutorials/rag-integration","title":"RAG Integration","description":"This tutorial guides you through setting up and configuring Solace Agent Mesh (SAM) Retrieval Augmented Generation (RAG) plugin. The RAG plugin enables your agents to answer questions by retrieving information from a knowledge base of your documents.","source":"@site/docs/documentation/tutorials/rag-integration.md","sourceDirName":"documentation/tutorials","slug":"/documentation/tutorials/rag-integration","permalink":"/solace-agent-mesh/docs/documentation/tutorials/rag-integration","draft":false,"unlisted":false,"editUrl":"https://github.com/SolaceLabs/solace-agent-mesh/edit/main/docs/docs/documentation/tutorials/rag-integration.md","tags":[],"version":"current","sidebarPosition":70,"frontMatter":{"title":"RAG Integration","sidebar_position":70,"toc_max_heading_level":4},"sidebar":"docSidebar","previous":{"title":"Slack Integration","permalink":"/solace-agent-mesh/docs/documentation/tutorials/slack-integration"},"next":{"title":"SAM Gateway Migration Guide: Upgrading to the A2A SDK","permalink":"/solace-agent-mesh/docs/documentation/migration-guides/a2a-upgrade-to-0.3.0/a2a-gateway-upgrade-to-0.3.0"}}');var a=t(4848),r=t(8453);const o={title:"RAG Integration",sidebar_position:70,toc_max_heading_level:4},s="RAG Integration",l={},c=[{value:"What is SAM RAG?",id:"what-is-sam-rag",level:2},{value:"Prerequisites",id:"prerequisites",level:2},{value:"Adding the RAG Plugin",id:"adding-the-rag-plugin",level:2},{value:"Configuring the RAG Agent",id:"configuring-the-rag-agent",level:2},{value:"Shared Configuration",id:"shared-configuration",level:3},{value:"RAG Pipeline Configuration",id:"rag-pipeline-configuration",level:3},{value:"1. Scanner Configuration",id:"1-scanner-configuration",level:4},{value:"2. Preprocessor Configuration",id:"2-preprocessor-configuration",level:4},{value:"3. Splitter Configuration",id:"3-splitter-configuration",level:4},{value:"4. Embedding Configuration",id:"4-embedding-configuration",level:4},{value:"5. Vector Database Configuration",id:"5-vector-database-configuration",level:4},{value:"6. LLM Configuration",id:"6-llm-configuration",level:4},{value:"7. Retrieval Configuration",id:"7-retrieval-configuration",level:4},{value:"Environment Variables",id:"environment-variables",level:3},{value:"Running the RAG Agent",id:"running-the-rag-agent",level:2},{value:"Testing the RAG Agent",id:"testing-the-rag-agent",level:2},{value:"Ingesting Documents",id:"ingesting-documents",level:3},{value:"Option 1: Automatic Scanning (Batch Ingestion)",id:"option-1-automatic-scanning-batch-ingestion",level:4},{value:"Option 2: Manual Upload via Gateway",id:"option-2-manual-upload-via-gateway",level:4},{value:"Querying the Knowledge Base",id:"querying-the-knowledge-base",level:3},{value:"Troubleshooting",id:"troubleshooting",level:2}];function d(e){const n={a:"a",blockquote:"blockquote",code:"code",h1:"h1",h2:"h2",h3:"h3",h4:"h4",header:"header",li:"li",ol:"ol",p:"p",pre:"pre",strong:"strong",ul:"ul",...(0,r.R)(),...e.components};return(0,a.jsxs)(a.Fragment,{children:[(0,a.jsx)(n.header,{children:(0,a.jsx)(n.h1,{id:"rag-integration",children:"RAG Integration"})}),"\n",(0,a.jsx)(n.p,{children:"This tutorial guides you through setting up and configuring Solace Agent Mesh (SAM) Retrieval Augmented Generation (RAG) plugin. The RAG plugin enables your agents to answer questions by retrieving information from a knowledge base of your documents."}),"\n",(0,a.jsx)(n.h2,{id:"what-is-sam-rag",children:"What is SAM RAG?"}),"\n",(0,a.jsx)(n.p,{children:"The SAM RAG plugin enhances your agents with the ability to perform retrieval-augmented generation. This means the agent can:"}),"\n",(0,a.jsxs)(n.ul,{children:["\n",(0,a.jsxs)(n.li,{children:[(0,a.jsx)(n.strong,{children:"Scan"})," documents from various sources (local filesystem, Google Drive, etc.)."]}),"\n",(0,a.jsxs)(n.li,{children:[(0,a.jsx)(n.strong,{children:"Preprocess"})," and ",(0,a.jsx)(n.strong,{children:"split"})," the text into manageable chunks."]}),"\n",(0,a.jsxs)(n.li,{children:[(0,a.jsx)(n.strong,{children:"Embed"})," these chunks into vectors and store them in a vector database."]}),"\n",(0,a.jsxs)(n.li,{children:[(0,a.jsx)(n.strong,{children:"Retrieve"})," relevant chunks of text based on a user's query."]}),"\n",(0,a.jsxs)(n.li,{children:[(0,a.jsx)(n.strong,{children:"Generate"})," an answer using a large language model (LLM) augmented with the retrieved information."]}),"\n"]}),"\n",(0,a.jsx)(n.p,{children:"This allows you to build agents that can answer questions about your own private data."}),"\n",(0,a.jsx)(n.h2,{id:"prerequisites",children:"Prerequisites"}),"\n",(0,a.jsx)(n.p,{children:"Before you begin, ensure you have:"}),"\n",(0,a.jsxs)(n.ul,{children:["\n",(0,a.jsxs)(n.li,{children:[(0,a.jsx)(n.a,{href:"/solace-agent-mesh/docs/documentation/getting-started/installation",children:"Installed Solace Agent Mesh and the SAM CLI"}),"."]}),"\n",(0,a.jsxs)(n.li,{children:[(0,a.jsx)(n.a,{href:"/solace-agent-mesh/docs/documentation/getting-started/quick-start",children:"Created a new Solace Agent Mesh project"}),"."]}),"\n",(0,a.jsx)(n.li,{children:"Access to a vector database (for example, Qdrant, Chroma, and Pinecone)."}),"\n",(0,a.jsx)(n.li,{children:"Access to an LLM for generation and an embedding model."}),"\n",(0,a.jsx)(n.li,{children:"A directory with some documents for the agent to ingest."}),"\n"]}),"\n",(0,a.jsx)(n.h2,{id:"adding-the-rag-plugin",children:"Adding the RAG Plugin"}),"\n",(0,a.jsx)(n.p,{children:"To add the RAG plugin to your SAM project, run the following command:"}),"\n",(0,a.jsx)(n.pre,{children:(0,a.jsx)(n.code,{className:"language-sh",children:"sam plugin add my-rag-agent --plugin sam-rag\n"})}),"\n",(0,a.jsxs)(n.p,{children:["Replace ",(0,a.jsx)(n.code,{children:"my-rag-agent"})," with your preferred agent name. This command:"]}),"\n",(0,a.jsxs)(n.ul,{children:["\n",(0,a.jsxs)(n.li,{children:["Installs the ",(0,a.jsx)(n.code,{children:"sam-rag"})," plugin."]}),"\n",(0,a.jsxs)(n.li,{children:["Creates a new agent configuration file at ",(0,a.jsx)(n.code,{children:"configs/agents/my-rag-agent.yaml"}),"."]}),"\n"]}),"\n",(0,a.jsx)(n.h2,{id:"configuring-the-rag-agent",children:"Configuring the RAG Agent"}),"\n",(0,a.jsxs)(n.p,{children:["The RAG agent requires a detailed configuration. Open ",(0,a.jsx)(n.code,{children:"configs/agents/my-rag-agent.yaml"})," to configure the following sections:"]}),"\n",(0,a.jsx)(n.h3,{id:"shared-configuration",children:"Shared Configuration"}),"\n",(0,a.jsxs)(n.p,{children:["Like other agents, the RAG agent needs a connection to the Solace broker and a configured LLM. This is typically done in a ",(0,a.jsx)(n.code,{children:"shared_config.yaml"})," file."]}),"\n",(0,a.jsx)(n.pre,{children:(0,a.jsx)(n.code,{className:"language-yaml",children:"# configs/shared_config.yaml\nshared_config:\n - broker_connection: &broker_connection\n dev_mode: ${SOLACE_DEV_MODE, false}\n broker_url: ${SOLACE_BROKER_URL, ws://localhost:8080}\n broker_username: ${SOLACE_BROKER_USERNAME, default}\n broker_password: ${SOLACE_BROKER_PASSWORD, default}\n broker_vpn: ${SOLACE_BROKER_VPN, default}\n temporary_queue: ${USE_TEMPORARY_QUEUES, true}\n\n - models:\n general: &general_model\n model: ${LLM_SERVICE_GENERAL_MODEL_NAME}\n api_base: ${LLM_SERVICE_ENDPOINT}\n api_key: ${LLM_SERVICE_API_KEY}\n"})}),"\n",(0,a.jsx)(n.h3,{id:"rag-pipeline-configuration",children:"RAG Pipeline Configuration"}),"\n",(0,a.jsxs)(n.p,{children:["The RAG pipeline has several stages, each with its own configuration block within the ",(0,a.jsx)(n.code,{children:"app_config"})," section of your ",(0,a.jsx)(n.code,{children:"my-rag-agent.yaml"})," file."]}),"\n",(0,a.jsx)(n.h4,{id:"1-scanner-configuration",children:"1. Scanner Configuration"}),"\n",(0,a.jsx)(n.p,{children:"The scanner discovers documents to be ingested. You can configure it to scan a local filesystem or cloud sources."}),"\n",(0,a.jsx)(n.p,{children:(0,a.jsx)(n.strong,{children:"Local Filesystem Example:"})}),"\n",(0,a.jsx)(n.pre,{children:(0,a.jsx)(n.code,{className:"language-yaml",children:'scanner:\n batch: true\n use_memory_storage: true\n source:\n type: filesystem\n directories:\n - "/path/to/your/documents" # Important: Replace with your actual document directory path\n filters:\n file_formats:\n - ".txt"\n - ".pdf"\n - ".md"\n'})}),"\n",(0,a.jsxs)(n.p,{children:[(0,a.jsx)(n.strong,{children:"Multi-Cloud Source Example:"}),"\nYou can also configure multiple sources, including Google Drive, OneDrive, and S3."]}),"\n",(0,a.jsx)(n.pre,{children:(0,a.jsx)(n.code,{className:"language-yaml",children:'scanner:\n batch: true\n use_memory_storage: true\n sources:\n - type: filesystem\n directories: ["${LOCAL_DOCUMENTS_PATH}"]\n - type: google_drive\n credentials_path: "${GOOGLE_DRIVE_CREDENTIALS_PATH}"\n folders:\n - folder_id: "${GOOGLE_DRIVE_FOLDER_ID_1}"\n name: "Documents"\n recursive: true\n'})}),"\n",(0,a.jsx)(n.h4,{id:"2-preprocessor-configuration",children:"2. Preprocessor Configuration"}),"\n",(0,a.jsx)(n.p,{children:"The preprocessor cleans the text extracted from documents."}),"\n",(0,a.jsx)(n.pre,{children:(0,a.jsx)(n.code,{className:"language-yaml",children:"preprocessor:\n default_preprocessor:\n type: enhanced\n params:\n lowercase: true\n normalize_whitespace: true\n remove_urls: true\n preprocessors:\n pdf: \n type: document\n params:\n lowercase: true\n normalize_whitespace: true\n remove_non_ascii: true\n remove_urls: true\n"})}),"\n",(0,a.jsx)(n.h4,{id:"3-splitter-configuration",children:"3. Splitter Configuration"}),"\n",(0,a.jsx)(n.p,{children:"The splitter breaks down large documents into smaller chunks. Different splitters are available for different file types."}),"\n",(0,a.jsx)(n.pre,{children:(0,a.jsx)(n.code,{className:"language-yaml",children:"splitter:\n default_splitter:\n type: recursive_character\n params:\n chunk_size: 2048\n chunk_overlap: 400\n splitters:\n markdown:\n type: markdown\n params:\n chunk_size: 2048\n chunk_overlap: 400\n pdf:\n type: token\n params:\n chunk_size: 500\n chunk_overlap: 100\n"})}),"\n",(0,a.jsx)(n.h4,{id:"4-embedding-configuration",children:"4. Embedding Configuration"}),"\n",(0,a.jsx)(n.p,{children:"This section defines the model used to create vector embeddings from the text chunks."}),"\n",(0,a.jsx)(n.pre,{children:(0,a.jsx)(n.code,{className:"language-yaml",children:'embedding:\n embedder_type: "openai"\n embedder_params:\n model: "${OPENAI_EMBEDDING_MODEL}"\n api_key: "${OPENAI_API_KEY}"\n api_base: "${OPENAI_API_ENDPOINT}"\n normalize_embeddings: true\n'})}),"\n",(0,a.jsx)(n.h4,{id:"5-vector-database-configuration",children:"5. Vector Database Configuration"}),"\n",(0,a.jsx)(n.p,{children:"Configure the connection to your vector database where the embeddings are stored."}),"\n",(0,a.jsx)(n.p,{children:(0,a.jsx)(n.strong,{children:"Qdrant Example:"})}),"\n",(0,a.jsx)(n.pre,{children:(0,a.jsx)(n.code,{className:"language-yaml",children:'vector_db:\n db_type: "qdrant"\n db_params:\n url: "${QDRANT_URL}"\n api_key: "${QDRANT_API_KEY}"\n collection_name: "${QDRANT_COLLECTION}"\n embedding_dimension: ${QDRANT_EMBEDDING_DIMENSION}\n'})}),"\n",(0,a.jsx)(n.p,{children:(0,a.jsx)(n.strong,{children:"Chroma Example:"})}),"\n",(0,a.jsx)(n.pre,{children:(0,a.jsx)(n.code,{className:"language-yaml",children:'vector_db:\n db_type: "chroma"\n db_params:\n host: "${CHROMA_HOST}"\n port: "${CHROMA_PORT}"\n collection_name: "${CHROMA_COLLECTION}"\n'})}),"\n",(0,a.jsx)(n.h4,{id:"6-llm-configuration",children:"6. LLM Configuration"}),"\n",(0,a.jsx)(n.p,{children:"Configure the LLM that is used to generate answers based on the retrieved context."}),"\n",(0,a.jsx)(n.pre,{children:(0,a.jsx)(n.code,{className:"language-yaml",children:'llm:\n load_balancer:\n - model_name: "gpt-4o"\n litellm_params:\n model: "openai/${OPENAI_MODEL_NAME}"\n api_key: "${OPENAI_API_KEY}"\n api_base: "${OPENAI_API_ENDPOINT}"\n'})}),"\n",(0,a.jsx)(n.h4,{id:"7-retrieval-configuration",children:"7. Retrieval Configuration"}),"\n",(0,a.jsx)(n.p,{children:"This defines how many document chunks are retrieved to answer a query."}),"\n",(0,a.jsx)(n.pre,{children:(0,a.jsx)(n.code,{className:"language-yaml",children:"retrieval:\n top_k: 7\n"})}),"\n",(0,a.jsx)(n.h3,{id:"environment-variables",children:"Environment Variables"}),"\n",(0,a.jsxs)(n.p,{children:["The RAG agent relies heavily on environment variables. Here are some of the most important ones you'll need to set in your ",(0,a.jsx)(n.code,{children:".env"})," file:"]}),"\n",(0,a.jsx)(n.pre,{children:(0,a.jsx)(n.code,{className:"language-bash",children:'# Solace Connection\nSOLACE_BROKER_URL=ws://localhost:8080\nSOLACE_BROKER_VPN=default\nSOLACE_BROKER_USERNAME=default\nSOLACE_BROKER_PASSWORD=default\nNAMESPACE=my-org/dev\n\n# LLM and Embedding Models\nOPENAI_API_KEY="your-openai-api-key"\nOPENAI_API_ENDPOINT="your-openai-api-endpoint"\nOPENAI_MODEL_NAME="model name. E.g., gpt-4o"\nOPENAI_EMBEDDING_MODEL="embedding model name. E.g., text-embedding-3-small"\n\n# Vector Database (Qdrant example)\nQDRANT_URL="Qdrant url"\nQDRANT_API_KEY="API key"\nQDRANT_COLLECTION="my-rag-collection"\nQDRANT_EMBEDDING_DIMENSION=1536 # Depends on your embedding model\n\n# Scanner\nLOCAL_DOCUMENTS_PATH="./my_documents" # Relative path to your documents folder\n'})}),"\n",(0,a.jsxs)(n.p,{children:["Create a directory named ",(0,a.jsx)(n.code,{children:"my_documents"})," in your project root and place some text or markdown files inside it."]}),"\n",(0,a.jsx)(n.h2,{id:"running-the-rag-agent",children:"Running the RAG Agent"}),"\n",(0,a.jsx)(n.p,{children:"Once you have configured your agent and set the environment variables, you can run it:"}),"\n",(0,a.jsx)(n.pre,{children:(0,a.jsx)(n.code,{className:"language-sh",children:"sam run configs/agents/my-rag-agent.yaml\n"})}),"\n",(0,a.jsx)(n.p,{children:"When the agent starts, it begins scanning the documents in the configured source, processing and ingesting them into your vector database. This process may take some time, depending on the number and size of your documents."}),"\n",(0,a.jsx)(n.h2,{id:"testing-the-rag-agent",children:"Testing the RAG Agent"}),"\n",(0,a.jsx)(n.p,{children:"Once your agent is running, you can test its retrieval capabilities and ingest new documents."}),"\n",(0,a.jsx)(n.h3,{id:"ingesting-documents",children:"Ingesting Documents"}),"\n",(0,a.jsx)(n.p,{children:"There are two primary ways to ingest documents into your RAG agent's knowledge base:"}),"\n",(0,a.jsx)(n.h4,{id:"option-1-automatic-scanning-batch-ingestion",children:"Option 1: Automatic Scanning (Batch Ingestion)"}),"\n",(0,a.jsxs)(n.p,{children:["This method uses the configured ",(0,a.jsx)(n.code,{children:"scanner"})," component. The agent automatically ingests documents from the directories specified in your configuration upon startup."]}),"\n",(0,a.jsx)(n.p,{children:(0,a.jsx)(n.strong,{children:"Step 1: Create a Document"})}),"\n",(0,a.jsxs)(n.p,{children:["First, create a simple text file named ",(0,a.jsx)(n.code,{children:"sam_features.txt"})," and add some content to it. For example:"]}),"\n",(0,a.jsx)(n.pre,{children:(0,a.jsx)(n.code,{className:"language-text",children:"Solace Agent Mesh (SAM) is a powerful framework for building AI agents.\nKey features of SAM include:\n- A flexible plugin architecture.\n- Integration with various LLMs and vector databases.\n- Scalable gateways for different communication protocols.\n- An event-driven design based on Solace PubSub+.\n"})}),"\n",(0,a.jsx)(n.p,{children:(0,a.jsx)(n.strong,{children:"Step 2: Place the Document in the Scanned Directory"})}),"\n",(0,a.jsxs)(n.p,{children:['In the "Environment Variables" section, we configured ',(0,a.jsx)(n.code,{children:"LOCAL_DOCUMENTS_PATH"})," to point to a directory (e.g., ",(0,a.jsx)(n.code,{children:"./my_documents"}),")."]}),"\n",(0,a.jsxs)(n.p,{children:["Create this directory in your project's root folder if you haven't already, and move your ",(0,a.jsx)(n.code,{children:"sam_features.txt"})," file into it."]}),"\n",(0,a.jsx)(n.pre,{children:(0,a.jsx)(n.code,{className:"language-sh",children:"mkdir -p my_documents\nmv sam_features.txt my_documents/\n"})}),"\n",(0,a.jsx)(n.p,{children:(0,a.jsx)(n.strong,{children:"Step 3: Run the Agent to Trigger Ingestion"})}),"\n",(0,a.jsx)(n.p,{children:"If your agent is already running, you'll need to restart it to trigger the batch scan. If it's not running, start it now:"}),"\n",(0,a.jsx)(n.pre,{children:(0,a.jsx)(n.code,{className:"language-sh",children:"sam run configs/agents/my-rag-agent.yaml\n"})}),"\n",(0,a.jsx)(n.p,{children:"You will see logs indicating that the file is being processed. Once the agent is running and the initial scan is complete, the document is successfully ingested and ready for retrieval."}),"\n",(0,a.jsx)(n.h4,{id:"option-2-manual-upload-via-gateway",children:"Option 2: Manual Upload via Gateway"}),"\n",(0,a.jsx)(n.p,{children:"You can also ingest documents dynamically by uploading them directly through a gateway, like the Web UI. This is useful for adding single documents without restarting the agent. The RAG agent exposes a tool for this purpose."}),"\n",(0,a.jsx)(n.p,{children:(0,a.jsx)(n.strong,{children:"Step 1: Start the RAG Agent and Web UI"})}),"\n",(0,a.jsx)(n.p,{children:"Ensure both your RAG agent and the Web UI gateway are running."}),"\n",(0,a.jsx)(n.p,{children:(0,a.jsx)(n.strong,{children:"Step 2: Upload a Document in the Web UI"})}),"\n",(0,a.jsxs)(n.ol,{children:["\n",(0,a.jsxs)(n.li,{children:["Open the Web UI (usually at ",(0,a.jsx)(n.a,{href:"http://localhost:8000",children:"http://localhost:8000"}),", or check your gateway configuration for the correct URL) and start a chat with your RAG agent."]}),"\n",(0,a.jsx)(n.li,{children:"Use the file attachment button to select a document from your local machine."}),"\n",(0,a.jsxs)(n.li,{children:["Send a prompt along with the file, instructing the agent to ingest it. For example:","\n",(0,a.jsxs)(n.blockquote,{children:["\n",(0,a.jsx)(n.p,{children:'"Please ingest the attached document into your knowledge base."'}),"\n"]}),"\n"]}),"\n"]}),"\n",(0,a.jsxs)(n.p,{children:["The RAG agent uses its ",(0,a.jsx)(n.code,{children:"built-in"})," ingest_document tool to process the file you uploaded. The file goes through the same preprocessing, splitting, and embedding pipeline as the documents from the automatic scan."]}),"\n",(0,a.jsx)(n.p,{children:(0,a.jsx)(n.strong,{children:"Step 3: Confirm Ingestion"})}),"\n",(0,a.jsx)(n.p,{children:"After the agent confirms that the document has been ingested, you can immediately ask questions about its content."}),"\n",(0,a.jsx)(n.h3,{id:"querying-the-knowledge-base",children:"Querying the Knowledge Base"}),"\n",(0,a.jsx)(n.p,{children:"You can interact with your RAG agent through any gateway, such as the Web UI gateway."}),"\n",(0,a.jsxs)(n.ol,{children:["\n",(0,a.jsx)(n.li,{children:"Make sure you have a Web UI gateway running (or add one to your project)."}),"\n",(0,a.jsxs)(n.li,{children:["Open the Web UI (usually at ",(0,a.jsx)(n.code,{children:"http://localhost:8000"}),")."]}),"\n",(0,a.jsxs)(n.li,{children:["Start a conversation with ",(0,a.jsx)(n.code,{children:"my-rag-agent"}),"."]}),"\n",(0,a.jsx)(n.li,{children:"Ask a question related to the content of the documents you provided during the initial scan."}),"\n"]}),"\n",(0,a.jsx)(n.p,{children:"For example, if you have a document about product features, you could ask:"}),"\n",(0,a.jsxs)(n.blockquote,{children:["\n",(0,a.jsx)(n.p,{children:'"What are the key features of Product X?"'}),"\n"]}),"\n",(0,a.jsx)(n.p,{children:"The agent searches its knowledge base, finds the relevant information, and generates an answer based on the content of your documents."}),"\n",(0,a.jsx)(n.h2,{id:"troubleshooting",children:"Troubleshooting"}),"\n",(0,a.jsxs)(n.ul,{children:["\n",(0,a.jsxs)(n.li,{children:[(0,a.jsx)(n.strong,{children:"Connection Errors"}),": Double-check all your URLs, API keys, and credentials for your LLM and vector database."]}),"\n",(0,a.jsxs)(n.li,{children:[(0,a.jsx)(n.strong,{children:"Ingestion Issues"}),": Check the agent logs for errors during the scanning, preprocessing, or embedding stages. Ensure the file formats are supported and the paths are correct."]}),"\n",(0,a.jsxs)(n.li,{children:[(0,a.jsx)(n.strong,{children:"No Answers"}),": If the agent can't answer, it might be because the information is not in the documents, or the ",(0,a.jsx)(n.code,{children:"top_k"})," retrieval setting is too low."]}),"\n"]})]})}function h(e={}){const{wrapper:n}={...(0,r.R)(),...e.components};return n?(0,a.jsx)(n,{...e,children:(0,a.jsx)(d,{...e})}):d(e)}},8453:(e,n,t)=>{t.d(n,{R:()=>o,x:()=>s});var i=t(6540);const a={},r=i.createContext(a);function o(e){const n=i.useContext(r);return i.useMemo((function(){return"function"==typeof e?e(n):{...n,...e}}),[n,e])}function s(e){let n;return n=e.disableParentContext?"function"==typeof e.components?e.components(a):e.components\|\|a:o(e.components),i.createElement(r.Provider,{value:n},e.children)}}}]);

solace-agent-mesh 1.0.9__py3-none-any.whl → 1.3.0__py3-none-any.whl

Potentially problematic release.

solace-agent-mesh 1.0.9py3-none-any.whl → 1.3.0py3-none-any.whl