How do I install the Oncall Handoff skill?

Run npx skillmds@latest add caipe-io/oncall-handoff-2 in your terminal (requires Node.js), paste this page's agent-chat prompt into Claude, Cursor, or any MCP-connected agent, or download the SKILL.md file and copy it into your agent's skills directory.

Is Oncall Handoff safe to use?

This skill has not completed SkillMD's automated safety review yet. Capability flags: docs only. SkillMD never runs a skill's scripts for you; review the SKILL.md before installing.

Which AI agents work with Oncall Handoff?

This skill is tagged as working with Claude Code, Claude.ai, OpenAI Codex. SKILL.md is an open format, so most agents that read a skills directory can load it too.

Is Oncall Handoff free to use?

Yes. Installing skills from SkillMD is free, and the skill stays under its author's original license.

Who published Oncall Handoff?

caipe-io (@caipe-io) published this skill. Their other Agent Skills are listed on their SkillMD profile.

Oncall Handoff by caipe-io — Agent Skill

Name: Oncall Handoff
Rating: 4.5 (4 reviews)
Author: caipe-io

Oncall Handoff

Generate a comprehensive on-call handoff document by aggregating open incidents, ongoing issues, recent deployments, and systems to watch. Orchestrates PagerDuty, Jira, and ArgoCD agents. Use during on-call rotation changes or shift handoffs.

caipe-io Updated Sep 22, 2026

File contents

On-Call Handoff

Build a structured handoff document for the incoming on-call engineer by collecting data from PagerDuty, Jira, and ArgoCD.

Instructions

Phase 1: Current Incident State (PagerDuty Agent)

Active incidents - all triggered and acknowledged incidents
Recently resolved incidents - resolved in the last 24 hours (may recur)
Current on-call schedule - who is on-call now, who is taking over
Upcoming maintenance windows - scheduled within the next 48 hours

Phase 2: Ongoing Issues (Jira Agent)

Open incident tickets - Jira issues labeled `incident`, `outage`, `p0`, `p1`
Known issues - issues labeled `known-issue` or `workaround`
Recently closed issues - resolved in last 48 hours that may have follow-ups
Pending changes - tickets in "Ready for Deploy" or "In Review" status

Phase 3: Environment State (ArgoCD Agent)

Recent deployments - applications synced in the last 48 hours
Unhealthy applications - any OutOfSync, Degraded, or Unknown apps
Pending syncs - applications with pending changes not yet deployed
Recent rollbacks - any applications that were rolled back

Phase 4: Compile Handoff Document

Organize all data into a structured, scannable document with clear action items.

Output Format

```markdown

On-Call Handoff Document

Date: February 9, 2026 Outgoing: @engineer-a | Incoming: @engineer-b

Active Incidents (Action Required)

Incident	Service	Urgency	Duration	Status
INC-789	auth-service	High	2h 15m	Acknowledged

Systems to Watch

auth-service - Connection pool issue ongoing
payment-api - Hotfix deployed yesterday, watch for regression ```

Examples

"Generate an on-call handoff document"
"Prepare a shift handoff for the incoming on-call engineer"
"What should the next on-call person know about?"
"Summarize the current state of production for handoff"

Guidelines

Prioritize actionable information - what does the incoming engineer need to DO?
Include workarounds for known issues so the incoming engineer does not have to search
Mark items as "action required" vs "monitor" vs "FYI" for clear prioritization
Always include escalation contacts with team context
Keep the last 48 hours as the default lookback window for context
If no active incidents exist, say so explicitly (this is good news)

---name: oncall-handoff-2description: Generate a comprehensive on-call handoff document by aggregating open incidents, ongoing issues, recent deployments, and systems to watch. Orchestrates PagerDuty, Jira, and ArgoCD agents. Use during on-call rotation changes or shift handoffs.---# On-Call HandoffBuild a structured handoff document for the incoming on-call engineer by collecting data from PagerDuty, Jira, and ArgoCD.## Instructions### Phase 1: Current Incident State (PagerDuty Agent)1. **Active incidents** - all triggered and acknowledged incidents2. **Recently resolved incidents** - resolved in the last 24 hours (may recur)3. **Current on-call schedule** - who is on-call now, who is taking over4. **Upcoming maintenance windows** - scheduled within the next 48 hours### Phase 2: Ongoing Issues (Jira Agent)1. **Open incident tickets** - Jira issues labeled \`incident\`, \`outage\`, \`p0\`, \`p1\`2. **Known issues** - issues labeled \`known-issue\` or \`workaround\`3. **Recently closed issues** - resolved in last 48 hours that may have follow-ups4. **Pending changes** - tickets in "Ready for Deploy" or "In Review" status### Phase 3: Environment State (ArgoCD Agent)1. **Recent deployments** - applications synced in the last 48 hours2. **Unhealthy applications** - any OutOfSync, Degraded, or Unknown apps3. **Pending syncs** - applications with pending changes not yet deployed4. **Recent rollbacks** - any applications that were rolled back### Phase 4: Compile Handoff DocumentOrganize all data into a structured, scannable document with clear action items.## Output Format\`\`\`markdown## On-Call Handoff Document**Date**: February 9, 2026**Outgoing**: @engineer-a | **Incoming**: @engineer-b### Active Incidents (Action Required)| Incident | Service | Urgency | Duration | Status ||----------|---------|---------|----------|--------|| INC-789 | auth-service | High | 2h 15m | Acknowledged |### Systems to Watch1. **auth-service** - Connection pool issue ongoing2. **payment-api** - Hotfix deployed yesterday, watch for regression\`\`\`## Examples- "Generate an on-call handoff document"- "Prepare a shift handoff for the incoming on-call engineer"- "What should the next on-call person know about?"- "Summarize the current state of production for handoff"## Guidelines- Prioritize actionable information - what does the incoming engineer need to DO?- Include workarounds for known issues so the incoming engineer does not have to search- Mark items as "action required" vs "monitor" vs "FYI" for clear prioritization- Always include escalation contacts with team context- Keep the last 48 hours as the default lookback window for context- If no active incidents exist, say so explicitly (this is good news)

Frequently asked questions