Agents

Agents are the core building block of SwarmClaw. Each agent has a personality, provider, and set of capabilities.

Creating an Agent

Navigate to Agents in the sidebar and click New Agent.

Fields

FieldDescription
NameDisplay name (e.g. "Code Reviewer")
DescriptionShort summary of what this agent does
SoulPersonality text — defines voice, tone, and character
System PromptTechnical instructions for the agent
ProviderWhich LLM provider to use
ModelSpecific model from the provider
ToolsEnable integrations like browser
SkillsSelect discovered skills from ~/.claude/skills/

Soul vs System Prompt

The soul defines who the agent is — personality, voice, quirks. It's prepended before the system prompt:

[User Preferences]
[Soul — personality/voice]
[System Prompt — technical instructions]
[Orchestrator context — if applicable]

Example soul:

You are a sharp, witty code reviewer who catches bugs others miss. You explain issues with dry humor and always suggest fixes, never just point out problems.

The system prompt is for technical instructions: what tools to use, output format, constraints.

Default Agent

SwarmClaw ships with a default "Assistant" agent using Claude CLI. It's a general-purpose agent you can customize or replace.

Tools

Currently available agent tools:

  • browser — MCP browser automation (Playwright). For Claude CLI, this passes --mcp-config. For LangGraph agents, it spawns a Playwright MCP subprocess.

Skills

Skills are markdown instruction files discovered from ~/.claude/skills/. Each skill's content is appended to the system prompt when assigned to an agent. Use the Refresh button to re-scan for new skills.

Orchestrator Agents

Mark an agent as an Orchestrator to enable multi-agent routing. See Orchestration.

Note: Orchestrators cannot use the Claude CLI provider — they require LangGraph-compatible providers (Anthropic, OpenAI, Ollama, or custom).