Skip to content
Open
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
79 changes: 79 additions & 0 deletions docs/community/tools/strands-agentcore-tools.md
Original file line number Diff line number Diff line change
@@ -0,0 +1,79 @@
---
project:
pypi: https://pypi.org/project/strands-agentcore-tools/
github: https://github.com/cagataycali/strands-agentcore-tools
maintainer: cagataycali
service:
name: AWS Bedrock AgentCore
link: https://aws.amazon.com/bedrock/agentcore/
---

# strands-agentcore-tools

{{ community_contribution_banner }}

[strands-agentcore-tools](https://github.com/cagataycali/strands-agentcore-tools) provides agentic tools that enable AI agents to autonomously deploy, manage, and monitor themselves on AWS Bedrock AgentCore.

## Installation

```bash
pip install strands-agentcore-tools
```

## Usage

```python
from strands import Agent
from strands_agentcore_tools import configure, launch, invoke, status, logs

agent = Agent(
tools=[configure, launch, invoke, status, logs],
system_prompt="You can deploy yourself to AWS AgentCore."
)

# Agent deploys itself
agent("""
Deploy yourself to AWS:
1. Configure deployment
2. Launch to production
3. Check status
4. Show logs
""")
```

## Key Features

- **9 Tools**: configure, launch, invoke, agents, status, logs, memory, identity, session
- **Zero DevOps**: Build & deploy via CodeBuild with ARM64 containers
- **Self-Deploying Agents**: Agents can autonomously deploy themselves
- **Memory Management**: Support for STM (Short-Term Memory) and LTM (Long-Term Memory)
- **OAuth Integration**: Built-in identity provider management

## Available Tools

| Tool | Purpose |
|------|---------|
| `configure` | Generate Dockerfile, IAM roles, config YAML |
| `launch` | Build & deploy to AgentCore via CodeBuild |
| `invoke` | Execute deployed agent with payload |
| `agents` | List/get agent runtimes |
| `status` | Check agent health & endpoint status |
| `logs` | Query CloudWatch logs |
| `memory` | Manage AgentCore memories (STM/LTM) |
| `identity` | OAuth provider management |
| `session` | Stop active runtime sessions |

## Configuration

Requires AWS credentials with permissions for:
- `bedrock-agentcore:*`
- `ecr:*`
- `codebuild:*`
- `iam:*`
- `logs:*`

## Resources

- [PyPI Package](https://pypi.org/project/strands-agentcore-tools/)
- [GitHub Repository](https://github.com/cagataycali/strands-agentcore-tools)
- [AWS Bedrock AgentCore Docs](https://docs.aws.amazon.com/bedrock-agentcore/)
88 changes: 88 additions & 0 deletions docs/community/tools/strands-fun-tools.md
Original file line number Diff line number Diff line change
@@ -0,0 +1,88 @@
---
project:
pypi: https://pypi.org/project/strands-fun-tools/
github: https://github.com/cagataycali/strands-fun-tools
maintainer: cagataycali
service:
name: Creative Tools
link: https://github.com/cagataycali/strands-fun-tools
---

# strands-fun-tools

{{ community_contribution_banner }}

[strands-fun-tools](https://github.com/cagataycali/strands-fun-tools) provides creative and interactive tools for building unique AI agents - including vision, audio, cursor control, and more.

## Installation

```bash
# Base installation (human_typer only)
pip install strands-fun-tools

# With specific features
pip install "strands-fun-tools[cursor,clipboard,vision]"

# Everything
pip install "strands-fun-tools[all]"
```

## Usage

```python
from strands import Agent
from strands_fun_tools import human_typer, cursor, clipboard

agent = Agent(
tools=[human_typer, cursor, clipboard],
system_prompt="You can type like a human and control the cursor!"
)

agent("Type 'Hello World!' with excited emotion and then copy it to clipboard")
```

## Key Features

### Interaction Tools
- **human_typer**: Human-like typing with emotions and typos
- **cursor**: Mouse & keyboard automation (pyautogui)
- **clipboard**: Clipboard monitoring & history
- **dialog**: Interactive terminal prompts

### Vision Tools
- **screen_reader**: OCR-based screen monitoring
- **yolo_vision**: Real-time object detection (YOLOv8)
- **face_recognition**: Face detection via AWS Rekognition
- **take_photo**: Camera capture & burst mode

### Audio Tools
- **listen**: Background audio transcription (Whisper)

### Connectivity
- **bluetooth**: BLE device monitoring & GATT operations

### Games
- **chess**: Stockfish chess engine integration

### Utilities
- **utility**: Crypto, encoding, hashing, JSON/YAML
- **spinner_generator**: Custom loading animations
- **template**: Jinja2 template rendering

## Tool Reference

| Tool | Install Extra | Key Actions |
|------|---------------|-------------|
| human_typer | *(base)* | Type with emotions: calm, excited, thoughtful |
| cursor | `[cursor]` | move, click, drag, type_text, hotkey |
| clipboard | `[clipboard]` | start, read, write, get_history |
| screen_reader | `[vision]` | start, capture_once, find_element |
| yolo_vision | `[vision]` | start, detect_once, query_objects |
| listen | `[audio]` | start, stop, get_transcripts |
| bluetooth | `[bluetooth]` | scan_once, list_devices, read_characteristic |
| chess | `[chess]` | new_game, get_best_move, make_move |

## Resources

- [PyPI Package](https://pypi.org/project/strands-fun-tools/)
- [GitHub Repository](https://github.com/cagataycali/strands-fun-tools)
71 changes: 71 additions & 0 deletions docs/community/tools/strands-google.md
Original file line number Diff line number Diff line change
@@ -0,0 +1,71 @@
---
project:
pypi: https://pypi.org/project/strands-google/
github: https://github.com/cagataycali/strands-google
maintainer: cagataycali
service:
name: Google APIs
link: https://developers.google.com/apis-explorer
---

# strands-google

{{ community_contribution_banner }}

[strands-google](https://github.com/cagataycali/strands-google) provides universal Google API integration for Strands Agents, offering access to 200+ Google services including Gmail, Drive, Calendar, YouTube, Sheets, Docs, and more.

## Installation

```bash
pip install strands-google
```

## Usage

```python
from strands import Agent
from strands_google import use_google, gmail_send, gmail_reply

agent = Agent(tools=[use_google, gmail_send, gmail_reply])

# Send an email
agent("Send an email to friend@example.com saying hello")

# Search Gmail
agent("Find all unread emails from last week")

# List Google Drive files
agent("Show me my recent Drive files")
```

## Key Features

- **200+ Google API Integrations**: Gmail, Drive, Calendar, YouTube, Sheets, Docs, and more
- **Flexible Authentication**: OAuth 2.0, Service Accounts, and API Keys
- **Gmail Helpers**: Easy email sending and replying with automatic encoding
- **Dynamic Scopes**: Configure OAuth scopes on-the-fly
- **Discovery API Support**: Access any Google API automatically

## Configuration

```bash
# OAuth 2.0 (recommended for Gmail/Drive/Calendar)
export GOOGLE_OAUTH_CREDENTIALS=~/gmail_token.json

# Service Account (for GCP services)
export GOOGLE_APPLICATION_CREDENTIALS=~/service-key.json

# API Key (for public APIs)
export GOOGLE_API_KEY=your_api_key_here
```

Run the authentication helper:
```bash
python -m strands_google.google_auth
```

## Resources

- [PyPI Package](https://pypi.org/project/strands-google/)
- [GitHub Repository](https://github.com/cagataycali/strands-google)
- [Google APIs Explorer](https://developers.google.com/apis-explorer)
72 changes: 72 additions & 0 deletions docs/community/tools/strands-rendercv.md
Original file line number Diff line number Diff line change
@@ -0,0 +1,72 @@
---
project:
pypi: https://pypi.org/project/strands-rendercv/
github: https://github.com/cagataycali/strands-rendercv
maintainer: cagataycali
service:
name: RenderCV
link: https://rendercv.com
---

# strands-rendercv

{{ community_contribution_banner }}

[strands-rendercv](https://github.com/cagataycali/strands-rendercv) provides AI-powered professional CV generation combining Strands Agents with RenderCV for beautiful, typographically perfect resumes.

## Installation

```bash
pip install strands-rendercv

# macOS: Install LaTeX for PDF generation
brew install --cask mactex-no-gui
```

## Usage

```python
from strands import Agent
from strands_rendercv import render_cv

agent = Agent(tools=[render_cv])

# Create template → Edit → Generate
agent.tool.render_cv(action="create_template")
agent.tool.render_cv(action="validate", input_file="John_Doe_CV.yaml")
agent.tool.render_cv(action="render", input_file="John_Doe_CV.yaml")
```

### AI-Powered Generation

```python
agent("""
Read my LinkedIn from ~/linkedin.md and work log from ~/work-log.md.
Generate CV for Senior Engineer at Tech Corp.
Focus on: distributed systems, leadership, open source.
Use 'engineeringresumes' theme.
""")
```

## Key Features

- **AI-Assisted Content**: Generate CV content from your documents
- **Schema Validation**: Validate YAML before rendering
- **Multi-Format Output**: PDF, HTML, and Markdown
- **5 Production-Ready Themes**: engineeringresumes, classic, sb2nov, moderncv, and more
- **Template Library**: Example CVs for various roles

## Available Actions

| Action | Description |
|--------|-------------|
| `create_template` | Generate sample YAML template |
| `validate` | Check schema before rendering |
| `render` | Generate PDF + HTML + MD |
| `list_themes` | Show available themes |

## Resources

- [PyPI Package](https://pypi.org/project/strands-rendercv/)
- [GitHub Repository](https://github.com/cagataycali/strands-rendercv)
- [RenderCV Documentation](https://rendercv.com)