v0.1.1 (#3)

* feat(config): maximize Bedrock model configuration settings (#1)

- Set Claude 3.7 Sonnet as default model with max tokens (64000)
- Extend timeouts to 15 minutes (900s) to prevent API timeouts
- Add adaptive retries for improved reliability
- Enable thinking capability with 2048 token budget
- Add STRANDS_MAX_TOKENS and STRANDS_BUDGET_TOKENS env variables
- Document configuration options in README

Co-authored-by: Cagatay Cali <[email protected]>

* chore(release): bump version to 0.1.1 and update project descriptions (#2)

* feat(config): maximize Bedrock model configuration settings

- Set Claude 3.7 Sonnet as default model with max tokens (64000)
- Extend timeouts to 15 minutes (900s) to prevent API timeouts
- Add adaptive retries for improved reliability
- Enable thinking capability with 2048 token budget
- Add STRANDS_MAX_TOKENS and STRANDS_BUDGET_TOKENS env variables
- Document configuration options in README

* chore(release): bump version to 0.1.1 and update project descriptions

* Improve README.md to better reflect agent-building focus

* Improve README.md to better reflect agent-building focus

* Improve README.md to better reflect agent-building focus

---------

Co-authored-by: Cagatay Cali <[email protected]>

---------

Co-authored-by: ./c² <[email protected]>
Co-authored-by: Cagatay Cali <[email protected]>

Author: 3 people

README.md

@@ -1,35 +1,35 @@
-# Strands CLI
+# Strands Agent Builder
 
-> When terminal meets intelligence: AI at your fingertips.
-
-A minimalist Strands Agents-powered CLI assistant for your terminal. Strands brings the power of Claude 3.7 Sonnet to your command line with advanced tool integration and knowledge base capabilities.
+An interactive Strands agent toolkit designed to help you build, test, and extend your own custom AI agents and tools. With the Strands Agent Builder, you can create specialized agents, develop custom tools, and compose complex AI workflows—all from your terminal.
 
 ## Quick Start
 
 ```bash
 # Install
 pipx install strands-agents-builder
 
-# Run interactive mode
+# Run interactive mode for agent development
 strands
 
-# One-off query
-strands "What's the capital of France?"
+# Build a custom tool and use it immediately
+strands "Create a tool named sentiment_analyzer that analyzes text sentiment and test it with some examples"
 
-# Pipe content
-cat document.txt | strands "Summarize this"
+# Pipe content to build an agent based on specifications
+cat agent-spec.txt | strands "Build a specialized agent based on these specifications"
 
-# Use with knowledge base
-strands --kb YOUR_KB_ID "Tell me about our project"
+# Use with knowledge base to extend existing tools
+strands --kb YOUR_KB_ID "Load my previous calculator tool and enhance it with scientific functions"
 ```
 
 ## Features
 
+- 🏗️ Create and test custom tools with instant hot-reloading
+- 🤖 Build specialized agents with focused capabilities
+- 🔄 Extend existing tools and enhance their functionality
 - 💬 Interactive command-line interface with rich output
-- 🔍 One-off queries via arguments or pipes
 - 🛠️ Powerful integrated tools (12+ tools including shell, editor, HTTP, Python)
-- 🧠 Knowledge base integration for memory and context
-- 🎮 Customizable system prompt
+- 🧠 Knowledge base integration for persisting and loading tools
+- 🎮 Customizable system prompt for specialized agents
 - 🪄 Nested agent capabilities with tool delegation
 - 🔧 Dynamic tool loading for extending functionality
 - 🖥️ Environment variable management and customization
@@ -53,39 +53,87 @@ Strands comes with a comprehensive set of built-in tools:
 
 ## Knowledge Base Integration
 
-Strands can leverage Amazon Bedrock Knowledge Bases to retrieve information and remember conversations.
+Strands Agent Builder leverages Amazon Bedrock Knowledge Bases to store and retrieve custom tools, agent configurations, and development history.
 
 ```bash
-# Specify knowledge base via command line
-strands --kb YOUR_KB_ID "What does our API do?"
+# Load and extend tools from your knowledge base
+strands --kb YOUR_KB_ID "Load my data_visualizer tool and add 3D plotting capabilities"
 
 # Or set a default knowledge base via environment variable
 export STRANDS_KNOWLEDGE_BASE_ID="YOUR_KB_ID"
-strands "What were our key decisions last week?"
+strands "Find my most recent agent configuration and make it more efficient"
 ```
 
 Features:
-- 🔄 Automatic retrieval of relevant information before answering queries
-- 💾 Conversation storage for building persistent memory
-- 📝 Every exchange is stored with proper formatting for future reference
-- 🔍 Contextual awareness across multiple conversations
+- 🔄 Retrieve previously created tools and agent configurations
+- 💾 Persistent storage for your custom tools and agents
+- 🛠️ Ability to iteratively improve tools across sessions
+- 🔍 Find and extend tools built in previous sessions
 
 ## Nested Agent Capabilities
 
-Use the `strands` tool to create specialized sub-agents with their own tools and system prompts:
+Use the `strands` tool to prototype and test specialized sub-agents with their own tools and system prompts:
 
 ```python
-# Basic usage
-agent.tool.strand(query="List files in the current directory")
+# Create a specialized data analysis agent
+agent.tool.strand(
+    query="Build and test a data analysis agent",
+    tool_names=["python_repl", "editor", "http_request"],
+    system_prompt="You're an AI specialized in data analysis. Your task is to build tools for data processing and visualization."
+)
 
-# With specific tools
+# Create a tool-building agent focused on web automation
 agent.tool.strand(
-    query="Analyze this Python code",
-    tool_names=["python_repl", "editor"],
-    system_prompt="You are an expert Python developer specializing in code optimization."
+    query="Create a set of web automation tools for browser testing",
+    tool_names=["editor", "python_repl", "shell"],
+    system_prompt="You're an expert in creating web automation tools. Your specialty is developing reliable browser testing utilities."
 )
 ```
 
+## Model Configuration
+
+### Optimized Defaults
+
+Strands comes with optimized, maxed-out configuration settings for the Bedrock model provider:
+
+```json
+{
+    "model_id": "us.anthropic.claude-3-7-sonnet-20250219-v1:0",
+    "max_tokens": 64000,
+    "boto_client_config": {
+        "read_timeout": 900,
+        "connect_timeout": 900,
+        "retries": {
+            "max_attempts": 3,
+            "mode": "adaptive"
+        }
+    },
+    "additional_request_fields": {
+        "thinking": {
+            "type": "enabled",
+            "budget_tokens": 2048
+        }
+    }
+}
+```
+
+These settings provide:
+- Claude 3.7 Sonnet (latest high-performance model)
+- Maximum token output (64,000 tokens)
+- Extended timeouts (15 minutes) for complex operations
+- Automatic retries with adaptive backoff
+- Enabled thinking capability with 2,048 token budget for recursive reasoning
+
+You can customize these values using environment variables:
+
+```bash
+# Maximum tokens for responses
+export STRANDS_MAX_TOKENS=32000
+
+# Budget for agent thinking/reasoning
+export STRANDS_BUDGET_TOKENS=1024
+```
+
 ## Custom Model Provider
 
 You can configure strands to use a different model provider with specific settings by passing in the following arguments:
@@ -100,11 +148,7 @@ As an example, if you wanted to use the packaged Ollama provider with a specific
 strands --model-provider ollama --model-config '{"model_id": <ID>}'
 ```
 
-Strands is packaged with the following providers:
-| Name | Config |
-| ---- | ------ |
-| `bedrock` | [reference](<LINK>) |
-| `ollama` | [reference](<LINK>) |
+Strands Agent Builder is packaged with `bedrock` and `ollama`.
 
 If you have implemented a custom model provider ([instructions](<LINK>)) and would like to use it with strands, create a python module under the directory "$CWD/.models" and expose an `instance` function that returns an instance of your provider. As an example, assume you have:
 
@@ -122,7 +166,7 @@ You can then use it with strands by running:
 $ strands --model-provider custom_model --model-config <JSON|FILE>
 ```
 
-## Custom Systm Prompts
+## Custom System Prompts
 
 ```bash
 # Via environment variable

pyproject.toml

@@ -4,8 +4,8 @@ build-backend = "hatchling.build"
 
 [project]
 name = "strands-agents-builder"
-version = "0.1.0"
-description = "A minimal Strands Agents CLI"
+version = "0.1.1"
+description = "An example Strands agent demonstrating streaming, tool use, and interactivity from your terminal. This agent builder can help you to build your own agents and tools."
 readme = "README.md"
 requires-python = ">=3.10"
 license = {text = "Apache-2.0"}

src/strands_agents_builder/strands.py

@@ -39,8 +39,8 @@
 
 # Custom tools, handlers, utils
 from tools import (
-    strand,
     store_in_kb,
+    strand,
     welcome,
 )
 

src/strands_agents_builder/utils/model_utils.py

@@ -2,11 +2,30 @@
 
 import importlib
 import json
+import os
 import pathlib
 from typing import Any
 
+from botocore.config import Config
 from strands.types.models import Model
 
+# Default model configuration
+DEFAULT_MODEL_CONFIG = {
+    "model_id": "us.anthropic.claude-3-7-sonnet-20250219-v1:0",
+    "max_tokens": int(os.getenv("STRANDS_MAX_TOKENS", "64000")),
+    "boto_client_config": Config(
+        read_timeout=900,
+        connect_timeout=900,
+        retries=dict(max_attempts=3, mode="adaptive"),
+    ),
+    "additional_request_fields": {
+        "thinking": {
+            "type": "enabled",
+            "budget_tokens": int(os.getenv("STRANDS_BUDGET_TOKENS", "2048")),
+        }
+    },
+}
+
 
 def load_path(name: str) -> pathlib.Path:
     """Locate the model provider module file path.
@@ -37,10 +56,14 @@ def load_config(config: str) -> dict[str, Any]:
 
     Args:
         config: A JSON string or path to a JSON file containing model configuration.
+            If empty string or '{}', the default config is used.
 
     Returns:
         The parsed configuration.
     """
+    if not config or config == "{}":
+        return DEFAULT_MODEL_CONFIG
+
     if config.endswith(".json"):
         with open(config) as fp:
             return json.load(fp)

tools/rich_interface.py

@@ -1,7 +1,6 @@
 import textwrap
 from typing import Any
 
-from strands.types.tools import ToolResult, ToolUse
 from rich.console import Console
 from rich.markdown import Markdown
 from rich.panel import Panel
@@ -10,6 +9,7 @@
 from rich.table import Table
 from rich.text import Text
 from rich.tree import Tree
+from strands.types.tools import ToolResult, ToolUse
 
 TOOL_SPEC = {
     "name": "rich_interface",