AurelienClere-365AurelienClere-365/ado-cowork-plugin/skills/ado-search/SKILL.md

ado-search

Skill5 starsChanged 5 months agoMITView on GitHub

Use when the user wants to search across Azure DevOps — across code files, wiki pages, or work items — using keywords or natural language. "find all usages of TokenService in code", "search for ADR-007 in the wiki", "find work items mentioning Redis", "where is the connection string defined", "search for open bugs related to authentication", "find all files containing TODO comments", "search wiki for deployment guide", "find work items with tag security in the Platform project"

In this repository10 other files it gives its agents.
- ado-contextSkill
- ado-pipelinesSkill
- ado-pm-reportingSkill
- ado-project-discoverySkill
- ado-pull-requestsSkill
- ado-repositoriesSkill
- ado-sprints-capacitySkill
- ado-test-plansSkill
- ado-wikiSkill
- ado-work-itemsSkill

The file

We split this skill into the rules it gives an agent

5 rules. About 0 are common, 2 are distinctive, and 3 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.
3
Very specific
Tied to their project's names, files or tools. Probably skip.

Common
Distinctive
Very specific

Distinctive:
Search across Azure DevOps code repositories, wiki pages, and work items using keywords or natural language.
line 17
Distinctive:
When the user's intent spans multiple areas (e.g.
line 61
Very specific:
When doing impact analysis, use search_code first to find all call sites, then offer to create a work item tracking the change.
line 97
Very specific:
If activeProject is set, scope searches to that project by default.
line 105
Very specific:
If no context is set, search_code and search_wiki require a project — invoke Workflow E of ado-context to resolve it.
line 107

≈ “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 Search

Search across Azure DevOps code repositories, wiki pages, and work items using keywords or natural language. Returns ranked results with context snippets.

When to use

User wants to find code that references a specific class, method, or string
User wants to find a wiki page by keyword or topic
User wants to find work items matching a description or keyword (without writing WIQL)
User is doing an impact analysis and needs to find all references across repos

MCP tools used

Tool	Purpose
`search_code`	Full-text search across all code in the organisation's repositories
`search_wiki`	Full-text search across all wiki pages in the organisation
`search_workitem`	Full-text search across work items (title, description, comments)

Workflow

Code search

Extract the search term(s) from the user request.
Identify optional filters: project, repository, file extension, branch.
Call search_code(query, project, repo, fileExtension).
Group results by repository and file.
Present the Code Search Results table with file path, line number, and context snippet.

Wiki search

Extract the search term(s).
Identify optional filter: project.
Call search_wiki(query, project).
Present the Wiki Search Results table with page title, project, and excerpt.

Work item search

Extract the search term(s) or topic.
Identify optional filters: project, work item type, state.
Call search_workitem(query, project, type, state).
Present the Work Item Search Results table.

Combined search

When the user's intent spans multiple areas (e.g. "find everything related to Redis"), run all three tools in sequence and present results grouped by type (Code, Wiki, Work Items).

Output Format

Code Search Results

Repository	File	Line	Match
api-gateway	src/Auth/AuthService.cs	87	`var svc = new TokenService(config)`
identity-service	tests/TokenServiceTests.cs	23	`TokenService svc = fixture.Create()`
shared-lib	src/Interfaces/ITokenService.cs	5	`public interface ITokenService`

3 results across 3 repositories.

Wiki Search Results

Page	Project	Excerpt
Architecture/ADRs/ADR-007	Platform	"...decided to use Redis for session caching because..."
Operations/Runbooks/Cache	Platform	"...flush Redis cache via `redis-cli FLUSHDB`..."

Work Item Search Results

ID	Type	Title	State	Project
1042	Story	Add MFA to login flow	Active	Platform
1058	Bug	Session token not refreshed	New	Platform
1063	Task	Evaluate Redis Sentinel setup	Done	Platform

Notes

search_code searches the default branch of each repository unless a branch is specified.
Results from search_code are ranked by relevance; the most relevant files appear first.
For large result sets (>20 items), present the top 10 and offer to show more.
When doing impact analysis, use search_code first to find all call sites, then offer to create a work item tracking the change.
search_workitem is faster than constructing a WIQL query for keyword-based searches; use ado-work-items wit_query when the user needs precise field-based filtering.

Active context

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

If activeProject is set, scope searches to that project by default. For cross-project searches, the user can explicitly say "search across all projects" to override.
If no context is set, search_code and search_wiki require a project — invoke Workflow E of ado-context to resolve it. search_workitem can run org-wide without a project; mention this to the user if no context is set.

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