devduck 1.10.1

🦆 Self-adapting agent - one file.

🦆 DevDuck

One file. Self-healing. Builds itself as it runs.

An AI agent that hot-reloads its own code, fixes itself when things break, and expands capabilities at runtime. Terminal, browser, cloud — or all at once.

pipx install devduck && devduck

Watch the intro

---

What It Does

• Hot-reloads — edit source, agent restarts instantly
• Self-heals — errors trigger automatic recovery
• 60+ tools — shell, GitHub, browser control, speech, scheduler, ML, messaging
• Multi-protocol — CLI, TUI, WebSocket, TCP, MCP, IPC, Zenoh P2P
• Unified mesh — terminal + browser + cloud agents in one network
• Deploys anywhere — devduck deploy --launch → AWS AgentCore

Requirements: Python 3.10–3.13 + any model provider (AWS, Anthropic, OpenAI, Ollama, Gemini, etc.)

---

Quick Start

devduck                              # interactive REPL
devduck --tui                        # multi-conversation terminal UI
devduck "create a REST API"          # one-shot
devduck --record                     # record session for replay
devduck --resume session.zip         # resume from snapshot
devduck deploy --launch              # ship to AgentCore

import devduck
devduck("analyze this code")

---

Model Detection

Set your key. DevDuck figures out the rest.

export ANTHROPIC_API_KEY=sk-ant-...   # → uses Anthropic
export OPENAI_API_KEY=sk-...          # → uses OpenAI
export GOOGLE_API_KEY=...             # → uses Gemini
# or just have AWS credentials        # → uses Bedrock
# or nothing at all                   # → uses Ollama

Priority: Bedrock → Anthropic → OpenAI → GitHub → Gemini → Cohere → Writer → Mistral → LiteLLM → LlamaAPI → MLX → Ollama

Override: MODEL_PROVIDER=bedrock STRANDS_MODEL_ID=us.anthropic.claude-sonnet-4-20250514-v1:0 devduck

---

Tools

Runtime — no restart needed

manage_tools(action="add", tools="strands_fun_tools.cursor")
manage_tools(action="create", code='...')
manage_tools(action="fetch", url="https://github.com/user/repo/blob/main/tool.py")

Hot-reload from disk

Drop a .py file in ./tools/ → it's available immediately.

# ./tools/weather.py
from strands import tool
import requests

@tool
def weather(city: str) -> str:
    """Get weather for a city."""
    return requests.get(f"https://wttr.in/{city}?format=%C+%t").text

Static config

export DEVDUCK_TOOLS="strands_tools:shell,editor;devduck.tools:use_github,scheduler"

---

Architecture

devduck/
├── __init__.py       # the whole agent — single file
├── tui.py            # multi-conversation Textual UI
├── tools/            # 60+ built-in tools (hot-reloadable)
└── agentcore_handler.py  # AWS AgentCore deployment handler

User → [CLI/TUI/WS/TCP/MCP/IPC] → DevDuck Core → Tools → Response
                                        ↕                    ↕
                                   Zenoh P2P            Knowledge Base
                                        ↕
                                  Browser + Cloud
                                  (Unified Mesh)

Ports: 10000 (mesh relay) · 10001 (WebSocket) · 10002 (TCP) · 10003 (MCP)

TUI Concurrency Model

The TUI (devduck --tui) supports true concurrent conversations with shared awareness:

              ┌─────────────────────────────────┐
              │   SharedMessages (thread-safe)   │
              │   [msg1, msg2, msg3, msg4, ...]  │
              └───┬──────────┬──────────┬───────┘
                  │          │          │
                  ▼          ▼          ▼
            ┌──────────┐ ┌──────────┐ ┌──────────┐
            │ Agent #1 │ │ Agent #2 │ │ Agent #3 │
            │ cb → 🟦  │ │ cb → 🟩  │ │ cb → 🟨  │
            │ panel #1 │ │ panel #2 │ │ panel #3 │
            └──────────┘ └──────────┘ └──────────┘
            All agents share the SAME messages list

Each conversation creates a fresh Agent (like TCP/Telegram tools do), but all agents point their .messages at a single SharedMessages instance — a thread-safe list subclass that serializes all reads and writes via a lock. This gives you:

• True concurrency — separate Agent instances with separate callback handlers, no conflicts
• Real-time shared awareness — when Agent #1 appends a message, Agent #2 sees it immediately on its next loop iteration
• Correct ordering — the lock ensures messages are appended in the order they're produced
• Isolated rendering — each agent's callback handler routes streaming output to its own color-coded TUI panel

The shared history is capped at 100 messages (configurable via DEVDUCK_TUI_MAX_SHARED_MESSAGES) and auto-clears on context window overflow.

Comparison across interfaces:

Interface	Agent per request	Shared messages	Use case
CLI	No (reuse one)	N/A (single-threaded)	Sequential interactive REPL
TUI	Yes (fresh Agent)	Yes (SharedMessages)	Concurrent conversations with shared context
TCP	Yes (fresh DevDuck)	No (fully isolated)	External network clients
Telegram	Yes (fresh DevDuck)	No (fully isolated)	Chat bot, each user isolated
WebSocket	Yes (fresh DevDuck)	No (fully isolated)	Browser clients

---

Multi-Agent Networking

Zenoh P2P — zero config

# Terminal 1
devduck   # → Zenoh peer: hostname-abc123

# Terminal 2
devduck   # auto-discovers Terminal 1

zenoh_peer(action="broadcast", message="git pull && npm test")  # all peers
zenoh_peer(action="send", peer_id="hostname-abc123", message="status?")  # one peer

Cross-network: ZENOH_CONNECT=tcp/remote:7447 devduck

Unified Mesh — everything connected

Terminal DevDucks (Zenoh) + browser tabs (WebSocket) + AgentCore agents (AWS) share one ring context. Open mesh.html to join from a browser.

---

Deploy

devduck deploy --launch
devduck deploy --name reviewer --tools "strands_tools:shell,editor" --launch
devduck list        # see deployed agents
devduck invoke "analyze code" --name reviewer

---

Session Recording & Resume

devduck --record                  # captures sys/tool/agent events
devduck --resume session.zip      # restores conversation + state
devduck --resume session.zip "continue where we left off"

from devduck import load_session
session = load_session("session.zip")
session.resume_from_snapshot(2, agent=devduck.agent)

Asciinema: DEVDUCK_ASCIINEMA=true devduck → shareable .cast files.

---

Background Modes

# Standard ambient — thinks while you're idle
DEVDUCK_AMBIENT_MODE=true devduck

# Autonomous — works until done
🦆 auto
# Agent signals [AMBIENT_DONE] when finished

---

Messaging

# Telegram
TELEGRAM_BOT_TOKEN=... STRANDS_TELEGRAM_AUTO_REPLY=true devduck
telegram(action="start_listener")

# Slack
SLACK_BOT_TOKEN=xoxb-... SLACK_APP_TOKEN=xapp-... devduck
slack(action="start_listener")

# WhatsApp (via wacli, no Cloud API)
whatsapp(action="start_listener")

Each incoming message spawns a fresh DevDuck with full tool access.

---

MCP

Expose as server (Claude Desktop):

{"mcpServers": {"devduck": {"command": "uvx", "args": ["devduck", "--mcp"]}}}

Load external servers:

export MCP_SERVERS='{"mcpServers": {"docs": {"command": "uvx", "args": ["strands-agents-mcp-server"]}}}'

---

macOS

use_mac(action="calendar.events", days=7)
use_mac(action="mail.send", to="[e-mail]", subject="Hi", body="Hello")
use_mac(action="safari.read")
use_mac(action="system.screenshot", path="/tmp/shot.png")
use_mac(action="system.dark_mode", enable=True)
use_mac(action="keychain.get", service="MyApp", account="me")

apple_notes(action="list")
use_spotify(action="now_playing")

---

More Tools

Tool	What
shell	Interactive PTY shell
editor	File create/replace/insert/undo
use_github	GitHub GraphQL API
use_computer	Mouse, keyboard, screenshots
listen	Background Whisper transcription
lsp	Language server diagnostics
scheduler	Cron + one-time jobs
tasks	Parallel background agents
sqlite_memory	Persistent memory with FTS
dialog	Rich terminal UI dialogs
speech_to_speech	Nova Sonic / OpenAI / Gemini voice
rl	Train RL agents, fine-tune LLMs
scraper	HTML/XML parsing
use_agent	Nested agents with different models
retrieve / store_in_kb	Bedrock Knowledge Base RAG

All 60+ tools

Core: system_prompt · manage_tools · manage_messages · tasks · scheduler · sqlite_memory · dialog · notify · use_computer · listen · lsp · tui · session_recorder · view_logs

Network: tcp · websocket · ipc · mcp_server · zenoh_peer · agentcore_proxy · unified_mesh · mesh_registry · jsonrpc

Platform: use_mac · apple_notes · use_spotify · telegram · whatsapp · slack

Cloud: use_github · fetch_github_tool · gist · agentcore_config · agentcore_invoke · agentcore_logs · agentcore_agents · create_subagent · store_in_kb · retrieve

AI/ML: rl · speech_to_speech · use_agent · scraper

macOS Native: apple_nlp · apple_vision · apple_wifi · apple_sensors · apple_smc

Strands: shell · editor · file_read · file_write · calculator · image_reader · speak

---

AGENTS.md

Drop an AGENTS.md in your working directory. DevDuck auto-loads it into the system prompt. No config needed.

# AGENTS.md
## Project: My API
- Framework: FastAPI
- Tests: pytest
- Deploy: Docker on ECS

---

Config Reference

All environment variables

Variable	Default	What
MODEL_PROVIDER	auto	bedrock anthropic openai github gemini cohere writer mistral litellm llamaapi mlx ollama
STRANDS_MODEL_ID	auto	Model name
DEVDUCK_TOOLS	60+ tools	package:tool1,tool2;package2:tool3
DEVDUCK_LOAD_TOOLS_FROM_DIR	false	Auto-load ./tools/*.py
DEVDUCK_KNOWLEDGE_BASE_ID	—	Bedrock KB for auto-RAG
MCP_SERVERS	—	JSON config for MCP servers
DEVDUCK_ENABLE_WS	true	WebSocket server
DEVDUCK_ENABLE_ZENOH	true	Zenoh P2P
DEVDUCK_ENABLE_AGENTCORE_PROXY	true	Mesh relay
DEVDUCK_ENABLE_TCP	false	TCP server
DEVDUCK_ENABLE_MCP	false	MCP HTTP server
DEVDUCK_ENABLE_IPC	false	IPC socket
DEVDUCK_WS_PORT	10001	WebSocket port
DEVDUCK_TCP_PORT	10002	TCP port
DEVDUCK_MCP_PORT	10003	MCP port
DEVDUCK_AGENTCORE_PROXY_PORT	10000	Mesh relay port
DEVDUCK_AMBIENT_MODE	false	Background thinking
DEVDUCK_AMBIENT_IDLE_SECONDS	30	Idle before ambient starts
DEVDUCK_AMBIENT_MAX_ITERATIONS	3	Max ambient iterations
DEVDUCK_AUTONOMOUS_MAX_ITERATIONS	100	Max autonomous iterations
DEVDUCK_ASCIINEMA	false	Record .cast files
DEVDUCK_TUI_MAX_SHARED_MESSAGES	100	Max shared message history in TUI
DEVDUCK_LSP_AUTO_DIAGNOSTICS	false	Auto LSP after edits
TELEGRAM_BOT_TOKEN	—	Telegram bot
SLACK_BOT_TOKEN	—	Slack bot
SPOTIFY_CLIENT_ID	—	Spotify

---

Dev Setup

git clone [e-mail]:cagataycali/devduck.git && cd devduck
python3.13 -m venv .venv && source .venv/bin/activate
pip install -e . && devduck

---

GitHub Actions

- uses: cagataycali/devduck@main
  with:
    task: "Review this PR"
    provider: "github"
    model: "gpt-4o"
  env:
    GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}

---

Citation

@software{devduck2025,
  author = {Cagatay Cali},
  title = {DevDuck: Self-Modifying AI Agent},
  year = {2025},
  url = {https://github.com/cagataycali/devduck}
}

---

Apache 2.0 · Built with Strands Agents · @cagataycali