Package index • deputy

Agent

Core agent class for agentic AI workflows

Agent: Agent R6 Class
AgentResult: Agent Result R6 Class
AgentEvent(): Create an agent event

Tools

Built-in tools and tool bundles

tool_ask_user(): AskUserQuestion tool
tool_list_files(): List files in a directory
tool_read_csv(): Read a CSV file
tool_read_file(): Read file contents
tool_run_bash(): Execute bash commands
tool_run_r_code(): Execute R code
tool_web_fetch(): Fetch web page content
tool_web_search(): Search the web
tool_write_file(): Write content to a file
tools_all(): All built-in tools
tools_code(): Code execution tools
tools_data(): Data reading tools
tools_file(): File operation tools
tools_interactive(): Tools for interactive workflows
tools_mcp(): Get tools from MCP servers
tools_preset(): Get a tool preset by name
tools_web(): Web tools
ToolPresets: Available tool preset names
list_presets(): List available tool presets
set_ask_user_callback(): Set callback for non-interactive user input

MCP Integration

Model Context Protocol server support

mcp_available(): Check if MCP support is available
mcp_servers(): List available MCP servers

Permissions

Control what agents can do

Permissions: Permissions R6 Class
PermissionMode: Permission modes for agent tool access
PermissionResultAllow(): Create an allow permission result
PermissionResultDeny(): Create a deny permission result
permissions_full(): Create a full access permission policy
permissions_readonly(): Create a read-only permission policy
permissions_standard(): Create a standard permission policy

Hooks

Intercept and customize agent behavior

HookMatcher: HookMatcher R6 Class
HookRegistry: HookRegistry R6 Class
HookEvent: Hook events supported by deputy
HookResultPostToolUse(): Create a PostToolUse hook result
HookResultPreCompact(): Create a PreCompact hook result
HookResultPreToolUse(): Create a PreToolUse hook result
HookResultSessionEnd(): Create a SessionEnd hook result
HookResultSessionStart(): Create a SessionStart hook result
HookResultStop(): Create a Stop hook result
HookResultSubagentStop(): Create a SubagentStop hook result
hook_block_dangerous_bash(): Create a hook that blocks dangerous bash commands
hook_limit_file_writes(): Create a hook that limits file writes to a directory
hook_log_tools(): Create a hook that logs all tool calls

Skills

Extend agents with specialized capabilities

Skill: Skill R6 Class
skill_load(): Load a skill from a directory
skill_create(): Create a skill programmatically
skills_list(): List available skills in a directory

Multi-Agent

Coordinate multiple specialized agents

LeadAgent: LeadAgent R6 Class
agent_definition(): Create an Agent Definition
agent_with_delegation(): Create a simple delegation agent

Errors

Error classes and constructors

deputy-errors DeputyError: Deputy Error Classes
is_deputy_error(): Check if an object is a deputy error
abort_budget_exceeded(): Abort with a budget exceeded error
abort_deputy(): Abort with a structured deputy error
abort_hook(): Abort with a hook error
abort_permission_denied(): Abort with a permission denied error
abort_provider(): Abort with a provider error
abort_session_load(): Abort with a session load error
abort_session_save(): Abort with a session save error
abort_tool_execution(): Abort with a tool execution error
abort_turn_limit(): Abort with a turn limit error