From Product Owner Agent Team for Azure DevOps
Safe, shared procedure for reading and updating Azure DevOps work items via the ado-remote-mcp server. Enforces read → diff → approval → write so no item is mutated without explicit confirmation. Use for any backlog fetch, query, field update, hierarchy re-parent, tagging, or state change. Triggers on: read work item, fetch backlog, WIQL query, update work item, change field, re-parent, set iteration, tag work item, bulk update, write to ADO, Azure DevOps update.
How this skill is triggered — by the user, by Claude, or both
Slash command
/Product Owner Agent Team for Azure DevOps:ado-work-item-ops Provide work item IDs or a query scope (area path / iteration), and what you want to read or change.Provide work item IDs or a query scope (area path / iteration), and what you want to read or change.The summary Claude sees in its skill listing — used to decide when to auto-load this skill
Single source of truth for how every Product Owner agent reads from and writes to Azure DevOps. It centralizes the safety pattern so agents do not each restate it.
Single source of truth for how every Product Owner agent reads from and writes to Azure DevOps. It centralizes the safety pattern so agents do not each restate it.
Use the ado-remote-mcp MCP server (configured in .vscode/mcp.json). If the server is not connected or returns an auth error, stop and tell the user to sign in to the MCP server — do not guess IDs or fabricate data.
Every operation follows this order. Never skip a step.
### Proposed changes — Work Item <ID>: <title>
| Field | Before | After |
|-------|--------|-------|
| <field> | <current> | <proposed> |
Confirm to apply? (yes / adjust / cancel)
## Work Items — <scope>
| ID | Type | Title | State | Parent | Iteration | Tags |
|----|------|-------|-------|--------|-----------|------|
Common WIQL templates (backlog slice, orphans, stale, oversized, ready-for-sprint, completed-in-iteration) live in wiql-queries.md. Adapt the project/area/iteration tokens before running.
Provides behavioral guidelines to reduce common LLM coding mistakes, focusing on simplicity, surgical changes, assumption surfacing, and verifiable success criteria.
Searches, retrieves, and installs Agent Skills from prompts.chat registry using MCP tools like search_skills and get_skill. Activates for finding skills, browsing catalogs, or extending Claude.
npx claudepluginhub chrriedel/product-owner-skills --plugin Product Owner Agent Team for Azure DevOps