/stop
Stop a specific agent.
How it fires
How this command gets triggered: by you, by Claude, or both.
- Fires itselfClaude auto-loads it when your prompt matches the work.
- You can call itInvoke it directly when you want it.
- Slash command
/stop
Context preview
What this command does when you run it.
Stop a specific agent.
Command definition
stop.mdStop Agent
Stop a specific agent.
Usage
/ops/stop <agent-name>
Instructions
1. **Validate the agent name** - Check it exists using `mcp__trinity__get_agent`
2. **Check current status**:
- If already stopped, report that no action is needed
- If running, proceed to stop
3. **Stop the agent** using `mcp__trinity__stop_agent`
4. **Verify the stop**:
- Confirm agent status is "stopped"
5. **Report the result**:
## Agent Stop: {agent-name}
**Previous Status**: {running/stopped}
**Stop Result**: {success/already stopped}
**Final Status**: {stopped}
{Any notes}Safety Checks
- Do NOT stop trinity-system (yourself)
- If agent is in the middle of a task, warn that it will be interrupted
- Report any errors clearly
Arguments
The agent name should be provided after the command. For example:
- `/ops/stop research-agent`
- `/ops/stop content-production-writer`
Self-hosted AI Agents Platform supporting Claude Code, Codex, Gemini agents. Apache 2.0.
Repo: Abilityai/trinity
Other commands on trinity.
- /ask
Answer a strategic question using accumulated research
Open command - /briefing
Generate daily briefing from research findings
Open command - /opportunities
List and rank all identified opportunities
Open command - /request-research
Ask the researcher to investigate a specific topic
Open command - /research
Run a research cycle on trending topics
Open command - /status
Report on recent research activity and findings count
Open command

