SkillBase

SkillBase is the abstract base class for all agent skills. Skills are modular, reusable capabilities — such as weather lookup, web search, or calendar access — that can be added to any AgentBase agent with a single call to agent.addSkill().

Extend SkillBase to create custom skills. Metadata (name, description, version, required packages, required environment variables, multi-instance support) is declared as static class constants on the subclass — the SkillRegistry and SkillManager read these constants directly. You override getTools() to expose SWAIG tools and optionally override lifecycle and prompt methods.

For the catalog of built-in skills and their configuration parameters, see the Skills page.

Static Class Constants

Subclasses declare their identity and requirements by overriding these static members. The SkillRegistry reads them directly — no manifest object is exchanged.

SKILL_NAME

stringRequired

Unique skill name used for registration. Subclasses must override with a non-empty string; the constructor throws otherwise.

SKILL_DESCRIPTION

stringRequired

Human-readable description. Subclasses must override with a non-empty string; the constructor throws otherwise.

SKILL_VERSION

stringDefaults to 1.0.0

Semantic version string.

REQUIRED_PACKAGES

readonly string[]Defaults to []

NPM packages the skill depends on. Checked at load time by validatePackages().

REQUIRED_ENV_VARS

readonly string[]Defaults to []

Environment variables the skill requires. Checked at load time by validateEnvVars().

SUPPORTS_MULTIPLE_INSTANCES

booleanDefaults to false

When true, the same skill can be added to an agent multiple times with different configurations (distinguished by tool_name).

getParameterSchema()

static method

Returns metadata about all parameters the skill accepts. See getParameterSchema.

Constructor

constructor(config?: SkillConfig)

The skill name, description, and version come from the subclass’s static constants — not from constructor arguments.

config

SkillConfig

Optional configuration key-value pairs for the skill instance. Any swaig_fields entry is extracted and automatically merged into tool definitions.

Instance Properties

skillName

string

The registered name for this skill instance, copied from SKILL_NAME (readonly).

instanceId

string

Unique identifier for this skill instance (readonly).

config

SkillConfig

Configuration parameters passed to the constructor (protected).

swaigFields

Record<string, unknown>

SWAIG metadata extracted from config, automatically merged into tool definitions (readonly).

Methods

setup

Initialize the skill. Return false to fail-closed on missing config.

cleanup

Release resources when the skill is removed or the agent shuts down.

getTools

Return SWAIG tool definitions this skill exposes.

getDataMapTools

Return fully-built SWAIG function dicts (DataMap style).

getParameterSchema

Return metadata about all parameters the skill accepts.

getGlobalData

Return data to merge into the agent’s global data.

getHints

Return speech recognition hints for this skill.

getInstanceKey

Get the unique key used to track this skill instance.

getPromptSections

Return prompt sections for the agent.

getSkillData

Extract this skill’s namespaced data from raw request data.

getSkillNamespace

Get the namespace key used to store this skill’s data.

updateSkillData

Write data under this skill’s namespace into a FunctionResult.

isInitialized

Check whether the skill has been initialized.

markInitialized

Mark the skill as initialized.

validateEnvVars

Validate that required environment variables are set.

hasAllEnvVars

Check whether all required env vars are present.

validatePackages

Validate that required packages can be imported.

hasAllPackages

Check whether all required packages are importable.

getConfig

Look up a configuration value by key.

getAgent

Get the agent this skill is attached to.

setAgent

Attach the skill to an agent.

defineTool

Imperatively register a tool with this skill.

defineTool (protected)

protected defineTool(toolDef: SkillToolDefinition): void

Imperatively register a tool. Use from setup() when the tool shape depends on config evaluated at runtime. The method merges this.swaigFields into the definition and pushes it onto the internal dynamic-tools list; the default getTools() returns that list. Skills with a static tool list should override getTools() directly instead.

getConfig

getConfig<T = unknown>(key: string, defaultValue?: T): T

Read a configuration value by key, with an optional fallback default.

key

stringRequired

The configuration key to look up.

defaultValue

Value to return if the key is not present in the config.

hasAllEnvVars / hasAllPackages

hasAllEnvVars(): boolean
async hasAllPackages(): Promise<boolean>

Boolean wrappers around validateEnvVars() and validatePackages().

Examples

Custom skill

import { SkillBase, type SkillToolDefinition } from '@signalwire/sdk';
class MyWeatherSkill extends SkillBase {
  static override SKILL_NAME = 'my_weather';
  static override SKILL_DESCRIPTION = 'Look up weather';
  static override SKILL_VERSION = '1.0.0';
  override getTools(): SkillToolDefinition[] {
    return [{
      name: 'get_weather',
      description: 'Get current weather for a city',
      parameters: { city: { type: 'string', description: 'City name' } },
      handler: async (args) => ({ response: 'Sunny, 72F' }),
    }];
  }
}

Using a skill

import { AgentBase } from '@signalwire/sdk';
const agent = new AgentBase({ name: 'weather-agent' });
agent.setPromptText("You are a helpful assistant.");
agent.addSkill("weather");
// Or with custom configuration
agent.addSkill("weather", { units: "celsius" });
agent.run();