Agent Zero Development Guide

This skill provides comprehensive, accurate guidance for extending and building features for Agent Zero. Use it when you need to:

• Understand the architecture and project layout
• Create new Tools for agent capabilities
• Add Extensions to hook into the framework lifecycle
• Build API Endpoints for the Web UI
• Create Agent Profiles (subordinates) with custom prompts
• Understand and extend the Prompt System
• Create Skills (see the dedicated create-skill skill for the full wizard)
• Work with Projects and workspace configuration

Path convention: Throughout this guide, /a0/ refers to the framework root — this is /a0/ inside Docker, or your local repository root in development. All paths are relative to this root.

[!IMPORTANT] Plugins are the primary way to extend Agent Zero. Most new tools, extensions, and prompts should be packaged as plugins. For all plugin tasks (create, review, manage, debug, contribute), load the a0-plugin-router skill which routes to the appropriate specialist. This guide covers the underlying framework patterns that plugins build upon.

Related skills: a0-plugin-router (plugin tasks) | create-skill (skill creation wizard) | a0-create-plugin | a0-review-plugin | a0-manage-plugin | a0-contribute-plugin | a0-debug-plugin

---

Architecture Overview

Project Layout

/a0/                              # Framework root
├── agent.py                      # Core Agent + AgentContext + AgentConfig classes
├── initialize.py                 # Agent initialization logic
├── models.py                     # Model definitions
├── run_ui.py                     # Web UI entry point
│
├── tools/                        # Core tools (search, response, browser, etc.)
├── extensions/
│   ├── python/                   # Python lifecycle extensions
│   │   ├── <hook_point>/         # e.g., agent_init/, system_prompt/, etc.
│   │   │   └── _NN_name.py       # Numbered extension files
│   │   └── _functions/           # Implicit @extensible decorator extensions
│   └── webui/                    # JavaScript WebUI extensions
│       └── <hook_point>/         # e.g., json_api_call_before/
│           └── name.js
├── api/                          # Flask API endpoint handlers
├── helpers/                      # Framework utilities and base classes
│   ├── tool.py                   # Tool + Response base classes
│   ├── extension.py              # Extension base class + @extensible decorator
│   ├── api.py                    # ApiHandler base class
│   ├── files.py                  # File operations + prompt reading
│   ├── plugins.py                # Plugin system manager
│   ├── print_style.py            # Console output formatting
│   └── ...                       # Many more utility modules
│
├── prompts/                      # Core prompt fragments (system, tools, framework)
├── agents/                       # Agent profiles (subordinate specializations)
│   ├── default/                  # Base profile (inherited by others)
│   ├── agent0/                   # Main user-facing agent
│   ├── developer/                # Developer subordinate
│   ├── hacker/                   # Security subordinate
│   ├── researcher/               # Research subordinate
│   └── _example/                 # Example profile with tool + extension samples
│
├── plugins/                      # Core plugins (tools, extensions, prompts)
│   ├── _code_execution/          # Terminal/Python/Node.js execution
│   ├── _memory/                  # Persistent memory system
│   ├── _text_editor/             # File read/write/patch
│   ├── _model_config/            # LLM model selection
│   ├── _infection_check/         # Prompt injection safety
│   └── ...                       # More core plugins
│
├── skills/                       # Core skills (SKILL.md bundles)
├── knowledge/                    # Knowledge base files
├── webui/                        # Web UI frontend
├── docs/                         # Documentation
│
└── usr/                          # User-space (survives updates)
    ├── agents/                   # User-created agent profiles
    ├── plugins/                  # User-installed plugins
    ├── skills/                   # User-created skills
    ├── knowledge/                # User knowledge base files
    ├── extensions/               # Standalone user extensions (created on demand; prefer plugins instead)
    ├── projects/                 # Project workspaces (created on demand when user adds projects via UI)
    └── workdir/                  # Default working directory

Key Architecture Patterns

1. Plugin-first design — Most capabilities (tools, extensions, prompts) are delivered via plugins in /a0/plugins/ (core) or /a0/usr/plugins/ (user).
2. Extensions execute in numeric order — Files named _10_*.py, _20_*.py, etc. run sequentially within each hook point.
3. Tools inherit from Tool — All tools implement the execute() method returning a Response.
4. Shared AgentContext — Enables state persistence across agents in a conversation.
5. Async/await throughout — All tool execution, extensions, and API handlers are async.
6. Prompt fragments compose — System prompts are assembled from named fragments with includes and variable substitution.
7. Profile inheritance — Agent profiles inherit from default/ and override specific prompt fragments.
8. User-space separation — Everything under /a0/usr/ survives framework updates.

Agent Loop

The core execution cycle works as follows:

1. User message arrives (via UI or API)
2. System prompt assembly — prompt fragments are composed with includes and variable substitution
3. LLM call — the assembled prompt + conversation history is sent to the model
4. Response parsing — the framework parses the LLM response looking for JSON tool calls
5. Tool execution — if tool calls are found, each tool's execute() method is called and the result is appended to history
6. Loop continues — steps 3-5 repeat until the agent produces a response tool call (which ends the loop) or a loop limit is reached

Extensions fire at each stage (e.g., monologue_start, before_main_llm_call, tool_execute_before, etc.), allowing plugins to observe and modify behavior at every point.

---

Creating Tools

Tools are how agents interact with the world. Each tool inherits from the Tool base class.

Import Path

from helpers.tool import Tool, Response

Tool Base Class

# /a0/helpers/tool.py

@dataclass
class Response:
    message: str              # Text response shown to agent
    break_loop: bool          # True = stop agent message loop
    additional: dict[str, Any] | None = None  # Extra metadata for history

class Tool:
    def __init__(self, agent: Agent, name: str, method: str | None,
                 args: dict[str,str], message: str,
                 loop_data: LoopData | None, **kwargs) -> None:
        self.agent = agent
        self.name = name
        self.method = method   # For tools with sub-methods (e.g., "skills_tool:load")
        self.args = args
        self.loop_data = loop_data
        self.message = message

    async def execute(self, **kwargs) -> Response:
        pass  # Override this

    # Lifecycle hooks (called automatically):
    async def before_execution(self, **kwargs): ...
    async def after_execution(self, response: Response, **kwargs): ...

Where Tools Live

Location	Purpose
/a0/tools/	Core framework tools (search, response, call_subordinate, etc.)
/a0/plugins/<plugin>/tools/	Plugin-provided tools (code_execution, memory, text_editor)
/a0/agents/<profile>/tools/	Profile-specific tool overrides
/a0/usr/plugins/<plugin>/tools/	User plugin tools

Example: Creating a Tool

Based on the actual _example profile in /a0/agents/_example/tools/example_tool.py:

# my_tool.py
from helpers.tool import Tool, Response

class MyTool(Tool):
    async def execute(self, **kwargs):
        # Get arguments — kwargs contains the tool_args from the agent's JSON
        input_data = kwargs.get("input", "")

        # Do something
        result = f"Processed: {input_data}"

        # Return response
        return Response(
            message=result,        # Shown to the agent
            break_loop=False,      # Don't stop the agent loop
        )

[!IMPORTANT] Every tool needs a corresponding prompt fragment so the agent knows how to use it. Create a file named agent.system.tool.<tool_name>.md in the appropriate prompts/ directory. See the Prompt System section.

Tool Best Practices

• Always handle errors gracefully — return error messages in Response, don't crash
• Access agent context via self.agent.context
• Use self.method to support sub-methods (e.g., my_tool:action1, my_tool:action2)
• Use kwargs.get() to read arguments with defaults
• For long operations, use self.set_progress() or self.add_progress() to show status
• Access self.loop_data for loop state (iteration count, timing, etc.) — this is the LoopData instance passed during tool dispatch

---

Creating Extensions

Extensions hook into specific lifecycle points in the agent framework.

Import Path

from helpers.extension import Extension

Extension Base Class

class Extension:
    def __init__(self, agent: "Agent | None", **kwargs):
        self.agent: "Agent | None" = agent
        self.kwargs = kwargs

    def execute(self, **kwargs) -> None | Awaitable[None]:
        pass  # Override this — kwargs are hook-point-specific

Extensions can be sync or async. If execute() returns an Awaitable, the framework will await it automatically. The agent parameter is nullable because some hook points (like startup_migration or banners) fire before an agent exists.

Extension File Location

Extensions live in directories named by their hook point. The path structure is:

extensions/python/<hook_point>/_NN_name.py

Where _NN_ is a numeric prefix controlling execution order (e.g., _10_, _20_, _50_).

Source	Path
Core extensions	/a0/extensions/python/<hook_point>/
Plugin extensions	/a0/plugins/<plugin>/extensions/python/<hook_point>/
User extensions	/a0/usr/extensions/python/<hook_point>/
Agent profile extensions	/a0/agents/<profile>/extensions/<hook_point>/
User plugin extensions	/a0/usr/plugins/<plugin>/extensions/python/<hook_point>/

Python Extension Hook Points

Complete list of available hook points:

Hook Point	When It Fires	Common Use
agent_init	Agent is initialized	Load configs, set defaults
system_prompt	System prompt is being assembled	Inject prompt content
monologue_start	Agent monologue begins	Pre-processing, state setup
message_loop_start	Before message processing loop	Pre-loop setup
message_loop_prompts_before	Before prompt assembly in loop	Modify prompt inputs
message_loop_prompts_after	After prompt assembly in loop	Add context (memory recall lives here)
before_main_llm_call	Before the LLM API call	Modify prompts, add context
util_model_call_before	Before utility model calls	Modify utility prompts
response_stream	When response streaming begins	Initialize stream handlers
response_stream_chunk	Per response chunk received	Transform output, collect data
response_stream_end	Response streaming complete	Finalize, analyze full response
reasoning_stream	Reasoning/thinking stream begins	Monitor reasoning
reasoning_stream_chunk	Per reasoning chunk	Collect reasoning data
reasoning_stream_end	Reasoning stream complete	Analyze reasoning
tool_execute_before	Before a tool runs	Validation, logging, safety checks
tool_execute_after	After a tool runs	Post-process results
hist_add_before	Before adding to history	Modify history entries
hist_add_tool_result	After tool result added to history	Log tool results
message_loop_end	After message processing loop	Post-loop cleanup
monologue_end	Agent monologue complete	Memorization, cleanup
process_chain_end	Entire processing chain done	Final cleanup
job_loop	Background job loop tick	Periodic background tasks
error_format	Error is being formatted	Custom error messages
startup_migration	Framework startup	Data migrations
banners	Startup banners displayed	Add custom banners
embedding_model_changed	Embedding model changed	Reload vector stores (fired programmatically, not a directory-based hook)
user_message_ui	User message from UI	Pre-process user input
webui_ws_connect	WebSocket client connects	Session setup
webui_ws_disconnect	WebSocket client disconnects	Session cleanup
webui_ws_event	WebSocket event received	Handle custom WS events

The @extensible Decorator (Implicit Extension Points)

Any framework function decorated with @extensible automatically gets two extension points:

_functions/<module_path>/<qualname_path>/start
_functions/<module_path>/<qualname_path>/end

The path mapping converts Python module paths and qualified names using / separators:

• Module agent.py → agent
• Class method Agent.handle_exception → Agent/handle_exception
• Full path: _functions/agent/Agent/handle_exception/start

For nested modules like helpers.history, a method History.add would map to _functions/helpers/history/History/add/start.

For example, a function Agent.handle_exception in module agent creates:

• _functions/agent/Agent/handle_exception/start
• _functions/agent/Agent/handle_exception/end

Extensions in these directories receive a data dict with:

• data["args"] — positional args (mutable)
• data["kwargs"] — keyword args (mutable)
• data["result"] — set this to short-circuit the function
• data["exception"] — set to a BaseException to force-raise

This is used by plugins like _error_retry to wrap core agent methods.

WebUI Extensions (JavaScript)

Client-side extensions live under extensions/webui/<hook_point>/:

Hook Point	When It Fires
json_api_call_before	Before a JSON API request
json_api_call_after	After a JSON API response
fetch_api_call_before	Before a fetch API request
fetch_api_call_after	After a fetch API response
get_message_handler	Register custom message renderers
set_messages_before_loop	Before messages are rendered
set_messages_after_loop	After messages are rendered
webui_ws_push	WebSocket push to client

Example: Creating an Extension

Based on the actual _example profile in /a0/agents/_example/extensions/agent_init/_10_example_extension.py:

# extensions/python/agent_init/_15_my_extension.py
from helpers.extension import Extension

class MyExtension(Extension):
    async def execute(self, **kwargs):
        # Access the agent
        agent = self.agent
        context = agent.context

        # Extension logic — kwargs content depends on the hook point
        agent.agent_name = "CustomAgent" + str(agent.number)

Extension Execution Order

Extensions execute in numeric order based on filename prefix:

_10_first.py      # Runs first
_20_second.py     # Runs second
_50_third.py      # Runs third

Use 10-number increments to leave room for future extensions.

---

Creating API Endpoints

API endpoints serve the Web UI and external clients using Flask.

Import Path

from helpers.api import ApiHandler
from flask import Request, Response

ApiHandler Base Class

class ApiHandler:
    def __init__(self, app: Flask, thread_lock: ThreadLockType):
        self.app = app
        self.thread_lock = thread_lock

    # Override these class methods to configure behavior:
    @classmethod
    def requires_loopback(cls) -> bool: return False   # Restrict to localhost
    @classmethod
    def requires_api_key(cls) -> bool: return False     # Require API key
    @classmethod
    def requires_auth(cls) -> bool: return True         # Require auth session
    @classmethod
    def get_methods(cls) -> list[str]: return ["POST"]   # HTTP methods
    @classmethod
    def requires_csrf(cls) -> bool: return cls.requires_auth()  # CSRF protection

    # Implement this:
    async def process(self, input: dict, request: Request) -> dict | Response:
        pass

    # Utility: get or create an agent context
    def use_context(self, ctxid: str, create_if_not_exists: bool = True) -> AgentContext:
        ...

Where API Endpoints Live

Location	Purpose
/a0/api/	Core API endpoints
/a0/plugins/<plugin>/api/	Plugin API endpoints
/a0/usr/plugins/<plugin>/api/	User plugin API endpoints

Endpoints are auto-discovered by filename. The route is derived from the filename (e.g., my_endpoint.py -> /api/my_endpoint).

Example: API Endpoint

# api/my_endpoint.py
from helpers.api import ApiHandler
from flask import Request, Response
from agent import AgentContext

class MyEndpoint(ApiHandler):
    @classmethod
    def get_methods(cls) -> list[str]:
        return ["GET", "POST"]

    async def process(self, input: dict, request: Request) -> dict:
        param = input.get("param", "default")

        # Get or create agent context
        ctxid = input.get("context", "")
        context = self.use_context(ctxid)

        return {
            "result": f"processed {param}",
            "context": context.id,
        }

---

Creating Agent Profiles

Agent profiles define specialized subordinates with custom prompts and behaviors.

For a guided, step-by-step wizard (scope selection, agent.yaml schema, prompt overrides, tool/extension stubs, test checklist) use the dedicated /a0/skills/a0-create-agent/SKILL.md skill.

Profile Directory Structure

agents/<profile-name>/
+-- agent.yaml                    # Required: profile metadata
+-- prompts/                      # Optional: prompt overrides
|   +-- agent.system.main.role.md         # Role definition (most common override)
|   +-- agent.system.main.communication.md # Communication style
|   +-- agent.system.tool.<name>.md       # Tool-specific prompts
+-- tools/                        # Optional: profile-specific tools
|   +-- my_tool.py
+-- extensions/                   # Optional: profile-specific extensions
    +-- <hook_point>/
        +-- _NN_extension.py

agent.yaml Format

The actual format is simple YAML with only three fields:

title: Developer
description: Agent specialized in complex software development.
context: Use this agent for software development tasks, including writing code,
  debugging, refactoring, and architectural design.

Field	Purpose
title	Display name shown in UI and agent selection
description	Brief description of the agent's specialization
context	Instructions for when to delegate to this profile

[!NOTE] There is no model configuration, temperature, or allowed_tools in the profile YAML. agent.yaml contains only title, description, and context. Profile-specific Main/Utility model settings are managed by the _model_config plugin in usr/agents/<profile>/plugins/_model_config/config.json. Tool availability is controlled by plugin activation.

Where Profiles Live

Location	Purpose
/a0/agents/	Core profiles (default, agent0, developer, hacker, researcher)
/a0/usr/agents/	User-created profiles (survives updates)

Prompt Override Mechanism

Profiles inherit all prompts from the default/ profile. To customize behavior, place prompt files with the same name in your profile's prompts/ directory. The framework searches profile-specific prompts first, then falls back to the default.

The most common override is agent.system.main.role.md which defines the agent's role and specialization.

Example: Creating a Profile

# /a0/usr/agents/data-analyst/agent.yaml
title: Data Analyst
description: Agent specialized in data analysis, visualization, and statistical modeling.
context: Use this agent for data analysis tasks, creating visualizations, statistical
  analysis, and working with datasets in Python.

<!-- /a0/usr/agents/data-analyst/prompts/agent.system.main.role.md -->

## Your role
You are a specialized data analysis agent.
Your expertise includes:
- Python data analysis (pandas, numpy, scipy)
- Data visualization (matplotlib, seaborn, plotly)
- Statistical modeling and hypothesis testing
- SQL queries and database analysis
- Data cleaning and preprocessing

## Process
1. Understand the data and the question
2. Choose appropriate tools and methods
3. Execute analysis with code_execution_tool
4. Visualize results when applicable
5. Provide clear interpretation of findings

Reference: The _example Profile

The framework includes a complete example profile at /a0/agents/_example/ that demonstrates:

• Custom tool: /a0/agents/_example/tools/example_tool.py
• Custom extension: /a0/agents/_example/extensions/agent_init/_10_example_extension.py
• Tool prompt: /a0/agents/_example/prompts/agent.system.tool.example_tool.md
• Role prompt: /a0/agents/_example/prompts/agent.system.main.role.md

---

Prompt System

Agent Zero assembles system prompts from named fragments using includes and variable substitution.

Prompt File Naming Convention

Prompt files follow a dot-separated naming scheme:

agent.system.main.md            # Main system prompt (entry point)
agent.system.main.role.md       # Role definition
agent.system.main.communication.md  # Communication style
agent.system.tool.<name>.md     # Tool usage instructions
agent.system.tools.md           # Tools overview
agent.system.projects.main.md   # Project system
agent.system.secrets.md         # Secret handling
agent.system.skills.md          # Skills listing
agent.system.datetime.md        # Current date/time
agent.context.extras.md         # Context extras
fw.*.md                         # Framework messages (errors, hints, etc.)

Where Prompts Live

Location	Priority	Purpose
/a0/agents/<profile>/prompts/	Highest	Profile-specific overrides
/a0/usr/agents/<profile>/prompts/	High	User profile overrides
/a0/plugins/<plugin>/prompts/	Normal	Plugin-provided prompts
/a0/usr/plugins/<plugin>/prompts/	Normal	User plugin prompts
/a0/prompts/	Base	Core framework prompts

The framework searches directories in priority order and uses the first match found.

Include Mechanism

Prompts can include other fragments using double-brace include directives.

The syntax uses opening double-brace, the keyword, and closing double-brace:

Directive	Purpose
{{include "agent.system.main.role.md"}}	Include a named prompt fragment
{{include "agent.system.main.communication.md"}}	Include another fragment
{{include original}}	Include the same file from the next lower-priority directory

The include original directive is particularly useful for extending rather than fully replacing a prompt — your override can include the base version and add to it.

Variable Substitution

Prompts support {{variable_name}} placeholders that are replaced at render time with values passed from the framework or plugin configuration.

Conditional Blocks

Prompts support conditional rendering based on variables.

Reading Prompts in Code

# From within an Agent method:
content = self.read_prompt("fw.some_message.md", variable1="value1")

# From helpers:
from helpers.files import read_prompt_file
content = read_prompt_file("template.md", _directories=[...], var="value")

---

Creating Skills

Skills are reusable instruction bundles that the agent loads on demand via the skills_tool. Each skill lives in a directory containing a SKILL.md file with YAML frontmatter.

Location	Purpose
/a0/skills/	Core skills (shipped with framework)
/a0/usr/skills/	User-created skills (survives updates)

The agent interacts with skills through JSON tool calls:

{"tool_name": "skills_tool:list", "tool_args": {}}
{"tool_name": "skills_tool:load", "tool_args": {"skill_name": "my-skill"}}

For the complete skill creation wizard — including SKILL.md format, frontmatter fields, directory structure, best practices, and examples — load the create-skill skill.

---

Working with Projects

Projects provide isolated workspaces with custom configuration.

Projects are typically created and managed via the Web UI. The .a0proj/ directory and project.json are auto-generated when you create a project through the UI.

Project Structure

/a0/usr/projects/<project-name>/
+-- .a0proj/
|   +-- project.json          # Project configuration
|   +-- agents.json           # Per-project agent overrides
|   +-- variables.env         # Non-sensitive variables
|   +-- secrets.env           # Encrypted secrets
|   +-- memory/               # Project-specific memory
|       +-- index.faiss
|       +-- index.pkl
|       +-- embedding.json
+-- <project-files>/          # Your project files (working directory)

project.json Format

{
    "title": "My Project",
    "description": "Project description",
    "instructions": "Markdown instructions for the agent when this project is active",
    "color": "#3a86ff",
    "git_url": "",
    "memory": "own",
    "file_structure": {
        "enabled": true,
        "max_depth": 5,
        "max_files": 20,
        "max_folders": 20,
        "max_lines": 250,
        "gitignore": ".a0proj/\nvenv/\n**/__pycache__/\n**/node_modules/\n**/.git/\n"
    }
}

Field	Purpose
title	Display name
description	Brief description
instructions	Markdown injected into agent system prompt when project is active
color	UI accent color (hex)
git_url	Optional Git repository URL
memory	"own" for project-specific memory, or shared
file_structure	Controls the working directory tree shown to the agent

---

Plugin System Overview

Plugins are the primary extension mechanism in Agent Zero. A plugin can bundle tools, extensions, prompts, API endpoints, helpers, and UI components into a self-contained package.

For all plugin tasks — creating, reviewing, managing, contributing, or debugging plugins — load the a0-plugin-router skill, which routes to the appropriate specialist skill.

Core Plugins

The framework ships with these core plugins in /a0/plugins/:

Plugin	Purpose
_code_execution	Terminal, Python, Node.js code execution
_memory	Persistent vector memory system
_text_editor	File read/write/patch with line numbers
_model_config	LLM model selection and configuration
_browser	Direct browser automation and WebUI viewing
_infection_check	Prompt injection safety checks
_error_retry	Retry on critical exceptions
_email_integration	Email communication via IMAP/SMTP
_telegram_integration	Telegram bot integration
_chat_branching	Branch chats from any message
_promptinclude	Persistent behavioral rules (*.promptinclude.md)
_plugin_installer	Install plugins from ZIP/Git/Hub
_plugin_scan	Security scanning for plugins
_plugin_validator	Plugin manifest and code validation

---

Common Patterns Reference

Accessing Agent Context

# Shared across all agents in a conversation
context = self.agent.context
data = context.data  # dict-like shared state

# Store data
data["my_key"] = my_value

# Retrieve data
value = data.get("my_key", default)

Using File Helpers

from helpers import files

# File operations
content = files.read_file("path/to/file")
files.write_file("path/to/file", content)
exists = files.exists("path/to/file")

# Read and render a prompt file
content = files.read_prompt_file("template.md", _directories=[...], var="value")

Console Output

from helpers.print_style import PrintStyle

PrintStyle.hint("Informational message")
PrintStyle.warning("Warning message")
PrintStyle.error("Error message")
PrintStyle(font_color="#85C1E9").print("Custom styled output")

Error Handling

from helpers.tool import Response

try:
    result = await risky_operation()
except Exception as e:
    PrintStyle.error(f"Operation failed: {e}")
    return Response(message=f"Error: {e}", break_loop=False)

---

Development Workflow

When building features for Agent Zero:

1. Choose Your Extension Point

Want to...	Use
Add a new agent capability	Tool (in a plugin)
Hook into agent lifecycle	Extension (in a plugin)
Add Web UI functionality	API endpoint + WebUI extension
Create a specialized agent	Agent profile
Bundle reusable instructions	Skill
Package everything together	Plugin (recommended)

2. Develop in User Space

• New plugins -> /a0/usr/plugins/<name>/
• New profiles -> /a0/usr/agents/<name>/
• New skills -> /a0/usr/skills/<name>/
• New extensions -> /a0/usr/extensions/python/<hook_point>/

3. Test and Iterate

• Local dev: Run python run_ui.py (default port 50001 at http://localhost:50001)
• Docker: Restart the container or use the UI restart button; check logs with docker logs -f <container_name>
• Test with minimal input first
• Verify in the Web UI

4. Contributing

For contribution guidelines, see /a0/docs/contribution.md. For plugin contributions to the community Plugin Index, load the a0-contribute-plugin skill.

---

Best Practices

DO

• Use the plugin system for new features (see a0-create-plugin skill)
• Follow existing code patterns and conventions
• Write clear docstrings and comments
• Handle errors gracefully in tools and extensions
• Create prompt fragments for every tool (agent.system.tool.<name>.md)
• Develop in /a0/usr/ directories to survive updates
• Test with the _example profile as a reference
• Use from helpers.* imports (not from python.helpers.*)

DON'T

• Modify files in /a0/plugins/ or /a0/tools/ directly (use usr/ space)
• Hardcode paths or configuration values
• Skip creating prompt files for tools
• Ignore the plugin system (it's the intended extension mechanism)
• Mix sync and async code carelessly
• Access internal structures when helpers exist

---

Quick Reference: Key Files

File	Purpose
/a0/agent.py	Core Agent, AgentContext, AgentConfig classes
/a0/helpers/tool.py	Tool + Response base classes
/a0/helpers/extension.py	Extension base + @extensible decorator
/a0/helpers/api.py	ApiHandler base class
/a0/helpers/files.py	File ops + prompt reading
/a0/helpers/plugins.py	Plugin system manager
/a0/helpers/print_style.py	Console output formatting
/a0/agents/_example/	Reference example profile with tool + extension
/a0/prompts/agent.system.main.md	Main system prompt entry point