AI Agents

Spawn Claude Code agents with full tool access. Every agent runs as a real CLI process.

Agent Launcher#

Open the Agent Launcher with Ctrl+Shift+A. From here you can select a built-in agent preset, create a custom agent with your own system prompt, choose a model (Opus, Sonnet, Haiku, or Codex models for OpenAI-backed agents), and configure per-project MCP servers.

Built-in Agents#

DAEMON ships with six pre-configured agents:

AgentPurpose
DebugTrace errors and fix stack traces
Security AuditScan for vulnerabilities and exploits
Code ReviewReview code quality and patterns
GitCommit, branch, and PR management
Test RunnerWrite and run test suites
SolanaOn-chain interactions and program development

Each agent runs as an independent Claude Code process with full filesystem access, terminal capabilities, and tool use.

Custom Agents#

  • Custom system prompts to define exactly how the agent should behave
  • Model selection to choose between Opus, Sonnet, and Haiku based on your task
  • MCP configuration to assign project-level or global MCP servers to each agent
DAEMON discovers Claude agents from ~/.claude/agents at runtime and lets you import them from the Agent Launcher: a dozen-plus ship with the repo, and any agent you add there becomes available across projects.

MCP Management#

Toggle project-level and global MCP servers from the DAEMON Console or Settings. Changes are written to .mcp.json in the project root (project-level) and ~/.claude.json (global), with a restart indicator when configs change.

Agent Sessions#

Every agent session is tracked in the Sessions tab on the right panel. Session data includes start/end timestamps, commands executed, files modified, and token usage.

For Solana projects, sessions can optionally be recorded on-chain via the Session Registry.