Use this when you want AgentPool to inspect Claude Code readiness, usage, and MCP configuration.
AgentPool will not log in to Claude, store credentials, scrape browser pages, or edit Claude config files.
agentpool setup claude-codeMachine-readable output:
agentpool setup claude-code --jsonSkip live usage probing:
agentpool setup claude-code --skip-usageThe setup checks:
- whether
claudeis installed; - model defaults and model catalog metadata;
- native Claude usage probing where available;
- Claude Code MCP install commands for AgentPool.
Preferred:
agentpool mcp-config --client claude-code --absolute-command --installRun one printed command. Scope choices:
local— personal to the current project (Claude default)project— team-shared.mcp.jsonin the repouser— all projects via~/.claude.json
Examples:
claude mcp add --transport stdio --scope local agentpool -- agentpool mcp
claude mcp add --transport stdio --scope project agentpool -- agentpool mcp
claude mcp add --transport stdio --scope user agentpool -- agentpool mcpThen verify:
claude mcp listInside Claude Code, run /mcp. Claude should report:
agentpool: .../agentpool mcp - ✓ Connected
See docs/mcp-clients.md for manual .mcp.json paste and
other hosts.
agentpool usage-summary --provider claude-code --refresh --json
agentpool models --provider claude-code
agentpool spawn \
--provider claude-code \
--model claude-opus-4-8 \
--reasoning-effort high \
--repo . \
--task "Inspect this repo read-only and summarize the main entry points." \
--isolation read_onlyClaude Code accepts aliases such as sonnet and opus, full names such as
claude-opus-4-8, and supported [1m] suffixes such as
claude-opus-4-8[1m]. Availability of 1M context depends on the Claude account
and plan; AgentPool only forwards the explicit model id.