agentleFS
Sign inStart free
AurelienClere-365AurelienClere-365/ado-cowork-plugin/skills/ado-pull-requests/SKILL.md

ado-pull-requests

Skill5 starsChanged 5 months agoMITView on GitHub

Use when the user asks about Azure DevOps pull requests, code reviews, PR comments, or wants to create, approve, or complete a PR. "show me my open pull requests", "list PRs waiting for my review", "approve PR #89", "add a review comment to PR #91", "create a pull request from feature/redis to main", "resolve the thread on PR #89", "show PR status and merge conflicts", "who reviewed PR #77", "complete PR #89 with squash merge", "add Alice as reviewer to PR #91"

The file

We split this skill into the rules it gives an agent

6 rules. About 0 are common, 2 are distinctive, and 4 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈2

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 4

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Call repo_pull_request(repo, prId) — returns title, description, source/target branches, reviewers, status, merge status, and thread count.

    line 49
  2. Distinctive:

    Only confirm write operations (create, approve, complete) with the user before executing.

    line 100
  3. Distinctive:

    Read-only operations (list, get) can proceed without confirmation.

    line 101
  4. Very specific:

    When the user says "approve PR #89 with comment", call repo_pull_request_write for the vote and repo_pull_request_thread_write for the …

    line 102
  5. Very specific:

    If activeProject is set, use it automatically as the project parameter — do not ask the user to re-specify it.

    line 112
  6. Very specific:

    If no context is set and a project is required, invoke Workflow E of ado-context to resolve it (list projects, ask user to pick, then …

    line 114

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

ADO Pull Requests

List, create, review, comment on, and complete Azure DevOps pull requests. Manage review threads and track the full PR lifecycle from draft to merge.

When to use

  • User wants to see open PRs (all, assigned to them, or authored by them)
  • User wants to create a new PR
  • User wants to approve, reject, or add reviewers to a PR
  • User wants to add a review comment or reply to a thread
  • User wants to resolve or reactivate a review thread
  • User wants to complete (merge) or abandon a PR

MCP tools used

ToolPurpose
repo_pull_requestList or get pull requests
repo_pull_request_writeCreate, update, approve, complete, or abandon a PR
repo_pull_request_threadList review threads and comments on a PR
repo_pull_request_thread_writeAdd, reply to, or update a review thread

Workflow

Listing PRs

  1. Identify filters: project, repository, status (active/completed/abandoned), reviewer (me), or author.
  2. Call repo_pull_request(project, repo, status, reviewerId).
  3. Present the PR List table.

Getting a single PR

  1. Call repo_pull_request(repo, prId) — returns title, description, source/target branches, reviewers, status, merge status, and thread count.
  2. Optionally call repo_pull_request_thread(prId) to show open review threads.

Creating a PR

  1. Extract: repo, source branch, target branch, title, description.
  2. Extract optional: reviewers, work item links, draft status.
  3. Confirm with the user (this is a write operation).
  4. Call repo_pull_request_write(create, ...).
  5. Confirm with the new PR ID and URL.

Approving / voting on a PR

  1. Identify PR ID and vote (approve, approve with suggestions, wait for author, reject).
  2. Call repo_pull_request_write(vote, prId, vote).
  3. Confirm the vote was recorded.

Adding a review comment

  1. Identify PR ID, file path and line (optional), and comment text.
  2. Call repo_pull_request_thread_write(prId, comment, filePath, line).
  3. Confirm the thread was created.

Completing a PR

  1. Identify PR ID and merge strategy (merge commit, squash, rebase, rebase+squash).
  2. Confirm with the user before proceeding — this is irreversible.
  3. Call repo_pull_request_write(complete, prId, mergeStrategy, deleteSourceBranch).
  4. Confirm with the merge commit hash.

Output Format

PR List

PR #TitleAuthorSource → TargetThreadsStatus
89feat: add Redis cache layerCarlos Silvafeature/redis → main3 openActive
91fix: null ref in OrderServiceAlice Martinfix/null-order → main0Draft
77chore: update dependenciesBob Nguyenchore/deps → main0Completed

Review Threads

ThreadFileLineStatusComments
#1src/Cache/RedisClient.cs45Active2 (Bob, Alice)
#2src/Cache/RedisClient.cs67Active1 (Bob)
#3README.md—Resolved1 (Carlos)

Notes

  • Only confirm write operations (create, approve, complete) with the user before executing. Read-only operations (list, get) can proceed without confirmation.
  • When the user says "approve PR #89 with comment", call repo_pull_request_write for the vote and repo_pull_request_thread_write for the comment — two tool calls.
  • Completing a PR that has conflicts will fail. Check mergeStatus from repo_pull_request first; if conflicts, inform the user and suggest resolving them before completing.
  • Draft PRs cannot be completed until published. Call repo_pull_request_write(publish) first if the user asks to complete a draft.

Active context

This skill respects the session context set by the ado-context skill.

  • If activeProject is set, use it automatically as the project parameter — do not ask the user to re-specify it.
  • If no context is set and a project is required, invoke Workflow E of ado-context to resolve it (list projects, ask user to pick, then proceed).

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

Check your setup, free