Power User Setup¶

A real-world .zshrc configuration for daily driving DevDuck with all features enabled.

Full Configuration¶

# Model — Claude Opus via Bedrock bearer token (fastest auth, no STS calls)
export AWS_BEARER_TOKEN_BEDROCK="ABSK..."
export STRANDS_MODEL_ID="global.anthropic.claude-opus-4-6-v1"
export STRANDS_MAX_TOKENS="64000"

# Tools — curated toolset (loads faster than all 60+)
export DEVDUCK_TOOLS="devduck.tools:use_github,editor,system_prompt,store_in_kb,manage_tools,websocket,zenoh_peer,agentcore_proxy,manage_messages,sqlite_memory,dialog,listen,use_computer,tasks,scheduler,telegram;strands_tools:retrieve,shell,file_read,file_write,use_agent"

# Knowledge Base — automatic RAG (stores & retrieves every conversation)
export STRANDS_KNOWLEDGE_BASE_ID="YOUR_KB_ID"

# MCP — auto-load Strands docs server
export MCP_SERVERS='{"mcpServers":{"strands-docs":{"command":"uvx","args":["strands-agents-mcp-server"]}}}'

# Messaging — Telegram & Slack bots
export TELEGRAM_BOT_TOKEN="your-telegram-bot-token"
export SLACK_BOT_TOKEN="xoxb-your-slack-bot-token"
export SLACK_APP_TOKEN="xapp-your-slack-app-token"

# Spotify control
export SPOTIFY_CLIENT_ID="your-client-id"
export SPOTIFY_CLIENT_SECRET="your-client-secret"
export SPOTIFY_REDIRECT_URI="http://127.0.0.1:8888/callback"

# Gemini as fallback/sub-agent model
export GEMINI_API_KEY="your-gemini-key"

What This Gives You¶

graph TB
    subgraph Config["⚙️ Power User Config"]
        Model["🧠 Claude Opus<br/>Bedrock Bearer Token"]
        KB["📚 Auto-RAG<br/>Knowledge Base"]
        MCP["📖 Strands Docs<br/>MCP Server"]
        Msg["📱 Telegram + Slack + WhatsApp"]
        Music["🎵 Spotify"]
        Mesh["🔗 Zenoh P2P + Mesh"]
        Tools["💬 26 Tools Loaded"]
    end

    Model --> DevDuck["🦆 DevDuck"]
    KB --> DevDuck
    MCP --> DevDuck
    Msg --> DevDuck
    Music --> DevDuck
    Mesh --> DevDuck
    Tools --> DevDuck

    DevDuck --> Expand["🔧 Expandable to 60+<br/>via manage_tools"]

    style Config fill:#1a1a2e,stroke:#4a90d9,color:#fff
    style DevDuck fill:#f5a623,stroke:#333,color:#000
    style Model fill:#9b59b6,stroke:#333,color:#fff
    style KB fill:#e74c3c,stroke:#333,color:#fff
    style MCP fill:#3498db,stroke:#333,color:#fff
    style Msg fill:#2ecc71,stroke:#333,color:#fff
    style Mesh fill:#7ed321,stroke:#333,color:#000

Feature	Benefit
🧠 Opus on Bedrock	Bearer token auth — zero-latency, no STS calls
📚 Auto-RAG	Every conversation stored in KB, context retrieved before each query
📖 Strands docs	SDK documentation available as MCP tools
📱 Telegram + Slack + WhatsApp	Three messaging channels ready to go
🎵 Spotify	Music control via `use_spotify`
🔗 Zenoh + Mesh	Auto-enabled multi-terminal awareness
💬 26 tools	Curated set on startup, expandable to 60+ on demand

Curated Tool Breakdown¶

The DEVDUCK_TOOLS string above loads exactly these tools:

From `devduck.tools` (16 tools)¶

Tool	Purpose
`use_github`	GitHub GraphQL API
`editor`	File create/replace/insert/undo
`system_prompt`	Self-improvement via prompt management
`store_in_kb`	Store to Bedrock Knowledge Base
`manage_tools`	Runtime tool add/remove/create
`websocket`	WebSocket server
`zenoh_peer`	P2P auto-discovery
`agentcore_proxy`	Unified mesh relay
`manage_messages`	Conversation history management
`sqlite_memory`	Persistent memory with FTS
`dialog`	Interactive terminal dialogs
`listen`	Background Whisper transcription
`use_computer`	Mouse, keyboard, screenshots
`tasks`	Background parallel agents
`scheduler`	Cron + one-time jobs
`telegram`	Telegram bot

From `strands_tools` (5 tools)¶

Tool	Purpose
`retrieve`	Bedrock KB retrieval
`shell`	Interactive PTY shell
`file_read`	File reading with search
`file_write`	File writing
`use_agent`	Spawn sub-agents with different models

Load more on demand¶

# Add Apple tools
manage_tools(action="add", tools="devduck.tools.apple_vision")
manage_tools(action="add", tools="devduck.tools.apple_nlp")

# Add Slack
manage_tools(action="add", tools="devduck.tools.slack")

# Add from external packages
manage_tools(action="add", tools="strands_fun_tools.clipboard", install=True)

Tips¶

Bearer Token Auth

AWS_BEARER_TOKEN_BEDROCK is faster than STS-based auth. No credential chain resolution needed. Get it from Bedrock console → "Get API Key".

Multi-Model Workflows

Set GEMINI_API_KEY to use Gemini as a sub-agent model:

use_agent(
    prompt="summarize this",
    system_prompt="You are a summarizer",
    model_provider="gemini"
)

AGENTS.md

Drop an AGENTS.md in your project directory. DevDuck auto-loads it into the system prompt — project-specific instructions without any config.

Ambient Mode

Enable background thinking while you're idle:

export DEVDUCK_AMBIENT_MODE=true

Or type ambient in the REPL. Type auto for fully autonomous mode.