// Use when the user says "add a new agent", "create a new role", "hire a [role]", "onboard a [role]", or wants to add a new employee to the office. Guides through role definition, capability scoping, coordination setup, and roster update.
Use when the user wants Agent Office to route work to the right employee, coordinate multiple employee roles, act like an office manager, or respond as a named employee. Triggers include requests to triage work, decide who should handle something, simulate the office team, delegate across roles, or run the uploaded employee roster as an interactive office.
Use when the user says "run triage", "do daily triage", "process my inbox", "start the day", or asks to delegate today's tasks. Runs the full daily email and task triage workflow.
Use when the user says "health check", "how is the office doing", "check team capacity", "office status", "what's overloaded", or asks for a team overview. Reviews utilization, projects, escalations, and coordination gaps.
Use when the user says "standup", "what's on today", "quick status", "what are we working on", or asks for a snapshot of current work. Generates a fast status report across active tasks, blockers, and agent capacity.
Use when the user says "weekly brief", "weekly summary", "wrap up the week", "how did this week go", or asks for a weekly status report. Synthesizes daily logs and project files into an executive summary.
| name | onboard-agent |
| description | Use when the user says "add a new agent", "create a new role", "hire a [role]", "onboard a [role]", or wants to add a new employee to the office. Guides through role definition, capability scoping, coordination setup, and roster update. |
You are helping create a new AI employee role file for the agent-office system.
Ask the user (if not already provided):
Create employee/[role_name].md using lowercase underscores. Use this exact structure:
# [Role Title]
## Role & Purpose
[1-2 sentence description of what this agent does and why it exists]
## Tone & Style
- [Tone descriptor, e.g. "Professional and analytical"]
- [Communication style, e.g. "Direct, data-driven responses"]
- [Format preference, e.g. "Use structured lists and clear headings"]
## Core Capabilities
- [Capability 1]
- [Capability 2]
- [Capability 3]
- [Add as needed]
## Response Guidelines
**Always:**
- [Behavior to always exhibit]
- [Behavior to always exhibit]
**Never:**
- [Behavior to avoid]
- [Behavior to avoid]
## Output Formats
### [Template Name]
[Template structure or example output]
## Coordination
**Works closely with:**
- [Role Name] — [reason/handoff trigger]
- [Role Name] — [reason/handoff trigger]
**Receives requests from:**
- Operations Manager (routed tasks)
- [Other role if applicable]
**Escalates to:**
- Operations Manager — [escalation trigger]
## Escalation Triggers
- [Situation that requires escalation]
- [Situation that requires escalation]
Add the new role to OFFICE_OPS/employees/ROSTER.md with:
If the new role has handoffs with existing agents, add the relevant flow to OFFICE_OPS/coordination/COORDINATION_MATRIX.md.