Why teams connect Plugin Eval
Trooper agents connect to Plugin Eval through secure OAuth or API credentials synced to your private OpenClaw server.
Ask Codex to evaluate a plugin or skill, give you a full analysis of a named plugin such as game-studio, explain why it scored that way, show what to fix first, explain its token budget, measure real token usage, benchmark a plugin, or tell you what to run next. Plugin Eval keeps the path engineer-friendly: start with a natural chat request, then use the local `plugin-eval start` entrypoint or the routed workflow command it recommends.
Every action runs as a traced ticket on the unified board. Combine Plugin Eval with GitHub, Gmail, Slack, and 1,000+ other plugins in multi-agent workflows.
Autonomous execution
Trooper agents call Plugin Eval APIs directly β search, read, write, and report back on traced tickets.
Approval gates
Sensitive writes wait for your authorization. Agents propose; you approve; the system executes.
Multi-agent workflows
Combine Plugin Eval with GitHub, Gmail, Slack, and 1,000+ plugins on one OpenClaw-powered board.
Plugin Eval on autopilot.
PR review, CI fixes, and release trains β traced on the board with Plugin Eval and harness agents.
What agents do with Plugin Eval
Search and summarize
Agent queries Plugin Eval for context before drafting a response or next action.
Create and update records
Write back to Plugin Eval with traced changes and human approval on sensitive ops.
Cross-tool workflows
Chain Plugin Eval with email, CRM, or code tools in a single multi-step mission.
Connect Plugin Eval
Credentials save to your private AI server and sync to OpenClaw. Configure in Settings β Plugins after deploy.
- 01
Deploy a Trooper unit at app.trooper.so
- 02
Open Settings β Plugins and connect Plugin Eval via OAuth or API key
- 03
Assign the Plugin Eval plugin to the right agent on your roster
- 04
Delegate a task from Slack, chat, or the unified task board
