respond
[experimental] Investigate and respond to a production incident. Pulls context, finds similar past incidents, suggests solutions, and enables coordination. Forked-subagent flow may not have MCP access in all Claude Code contexts -- prefer /rootly:status + /rootly:brief for now.
Install
With your agent
Install OMGSkills skill "rootly-ai-labs/rootly-claude-plugin:skills/respond" into this agent's normal skills directory: use the OMGSkills MCP get_skill tool, install only the pinned_install commit and path, verify SKILL.md against pinned_install.skill_md_sha, and stop if MCP or the pin is unavailable. Do not overwrite or change existing skills without asking me.
Manual command
git clone https://github.com/Rootly-AI-Labs/rootly-claude-plugin /tmp/rootly-claude-plugin && ln -s /tmp/rootly-claude-plugin/skills/respond ~/.claude/skills/respondFrom README
Incident Response (experimental) Experimental: this skill uses context: fork to delegate to the incident-investigator agent. In some Claude Code contexts the forked subagent does not inherit the plugin's MCP tools, in which case the agent will stop and report rather than fall back to bash/curl. If that happens, use the inline alternatives: /rootly:brief — stakeholder summary /rootly:status — service health /rootly:oncall — current responders You are helping the user investigate and respond to a production incident. This runs in a forked context to keep incident data separate from the main coding session. Workflow Identify the Incident If $ARGUMENTS contains an incident reference: If it is already a UUID (36-char hex with hyphens), use it directly with mcprootlygetIncident. If it looks like a sequential reference (4460, #4460, INC-4460), resolve to UUID via MCP: Normalize it to the exact incident number format INC- (for example, 4460 becomes INC-4460).