New: 98,000+ skills indexed from GitHub repos with 100+ stars · every skill safety-checkedBrowse top skillsTop skills
Esc to closeBrowse all skills
Submit a skillBrowse skills
Submit a skillBrowse skills

Official agent skill

Kb Search

by Azure in Azure/azure-sdk-tools

Query the APIView Copilot knowledge base for guidelines, examples, and memories.

OfficialMITAuto-check passedKnowledge Management

Install Kb Search

skills CLI
$ npx skills add Azure/azure-sdk-tools --skill kb-search -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI
$ gh skill install Azure/azure-sdk-tools kb-search --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy
$ git clone --depth 1 https://github.com/Azure/azure-sdk-tools.git skills-src && mkdir -p .claude/skills && cp -r skills-src/packages/python-packages/apiview-copilot/.github/skills/kb-search .claude/skills/kb-search && rm -rf skills-src

Use ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.

Claude Code skills documentation · loads skills from .claude/skills/

Facts

Skill name
kb-search
GitHub stars
134
Token cost
~1.6k tokens
SKILL.md length
689 words
Files
1
Skills in repo
35
Repo updated
First seen
Licence
MIT

At a glance

Query the APIView Copilot knowledge base for guidelines, examples, and memories.

  • Works in 3 steps: Text Search (--text) → File Search (--path) → ID Lookup (--ids)
  • Search knowledge base
  • SKILL.md covers When to Use, How It Works, Defaults and Three Query Modes, plus 4 more sections
  • Calls python

What it does

Kb Search is an agent skill from Azure/azure-sdk-tools, published by the product's own GitHub organization. Query the APIView Copilot knowledge base for guidelines, examples, and memories. Use for: search KB, search knowledge base, find guideline, lookup guideline, search guidelines, search examples, search memories, KB query, knowledge base search, find rule, what does the KB say about.

Its SKILL.md is about 1.6k tokens, which your agent loads only when the skill is triggered. It is a single SKILL.md file with no bundled scripts.

It sits in Knowledge Management, covering Knowledge bases. It works with Microsoft Azure, Azure AI Search and Azure Cosmos DB. The repository describes itself as: Tools repository leveraged by the Azure SDK team. The licence is MIT.

When your agent uses it

  • Search knowledge base
  • Lookup guideline
  • Search guidelines
  • Search examples

Example prompts

  • “/kb-search”

Requirements

  • Python 3

Workflow steps

3 steps, taken from the step headings in SKILL.md.

  1. Text Search (--text)
  2. File Search (--path)
  3. ID Lookup (--ids)

What it can do on your machine

Read from SKILL.md and the folder at commit cc5ca24. It shows what the files ask for, not the result of running them.

  • Tool permissions

    Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

    From allowed-tools in the SKILL.md frontmatter.

  • Runs code

    Shell commands in SKILL.md call:

    • python

    From the folder's file list and the shell code blocks in SKILL.md.

  • Network

    No URLs in SKILL.md.

    From URLs in SKILL.md, links to its own repository left out.

  • Credentials

    Names no API keys, tokens, secrets or passwords.

    From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Kb Search loads about 1.6k tokens when it runs. Until then it costs about 73 tokens; SKILL.md has 689 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~73
When it runs · the whole SKILL.md, loaded when a task matches
~1.6k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Auto-check passed

The automated check found no risky patterns in SKILL.md.

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

The full file from Azure/azure-sdk-tools at commit cc5ca24, republished under its MIT licence (© Azure). 689 words, ~1,632 tokens.

Download SKILL.mdSave it as .claude/skills/kb-search/SKILL.md (or your agent's skills folder).
name
kb-search
description
Query the APIView Copilot knowledge base for guidelines, examples, and memories. Use for: search KB, search knowledge base, find guideline, lookup guideline, search guidelines, search examples, search memories, KB query, knowledge base search, find rule, what does the KB say about.
argument-hint
Language + query text (e.g. 'Python async') or knowledge base IDs

When to Use

  • Looking up Azure SDK design guidelines for a specific language
  • Finding examples (good or bad) related to an API pattern
  • Retrieving specific knowledge base items by ID
  • Understanding what RAG context the AI reviewer sees for a given query
  • Investigating what guidance exists for a topic (e.g. naming, async, pagination)

How It Works

The knowledge base contains three entity types linked in a graph:

  • Guidelines — Azure SDK design rules (e.g. "Methods returning collections should use paging")
  • Examples — Good/bad code snippets linked to guidelines
  • Memories — Lessons learned from reviewer feedback, linked to guidelines and examples

A search query hits the Azure AI Search index (semantic + vector search), then resolves all linked entities from Cosmos DB via breadth-first traversal. The result is the same RAG context the AI reviewer would see.

Defaults

Unless the user says otherwise, always apply these defaults:

  • Language: Required for all non---ids queries (both --text and --path). Use the language the user is asking about.
  • Output: Use --markdown for text/path queries (easier for the agent to read and summarize). --markdown is not allowed with --ids.
  • Query mode: Use --text for natural language queries. Use --path to search from a file. Use --ids only when the user provides specific knowledge base IDs.

Three Query Modes

1. Text Search (--text)

Requires --language. Searches the index semantically and returns linked guidelines, examples, and memories.

bash
python cli.py kb search -l <language> --text "<query>" --markdown
2. File Search (--path)

Requires --language. Reads query text from a file and searches the index.

bash
python cli.py kb search -l <language> --path <file> --markdown
3. ID Lookup (--ids)

Retrieves specific items by ID. No other flags allowed (no --language, --text, --markdown, --path).

bash
python cli.py kb search --ids <id1> <id2> ...

Running the Command

Run directly in a foreground terminal with a 60-second timeout (timeout: 60000). By default, let the terminal capture the output so the agent can read it directly. If the output appears truncated (terminal truncates at ~60 KB), re-run with redirection to a file and read it back:

powershell
python cli.py kb search -l <language> --text "<query>" --markdown | Out-File -Encoding UTF8 scratch/kb_output.md

Do not use > — it produces UTF-16 in PowerShell 5.1. Always use | Out-File -Encoding UTF8.

Then use read_file on scratch/kb_output.md to get the full results.

Text search:

bash
python cli.py kb search -l <language> --text "<query>" --markdown

File search:

bash
python cli.py kb search -l <language> --path <file> --markdown

ID lookup:

bash
python cli.py kb search --ids <id1> <id2>

After the command completes, read the terminal output and summarize the findings for the user.

Examples
bash
# Search Python guidelines about async
python cli.py kb search -l python --text "async client methods" --markdown

# Search TypeScript guidelines about naming
python cli.py kb search -l typescript --text "method naming conventions" --markdown

# Search Java guidelines about pagination
python cli.py kb search -l java --text "pagination list operations" --markdown

# Retrieve specific items by ID
python cli.py kb search --ids guideline-abc123 example-def456

# JSON output instead of markdown (for structured processing)
python cli.py kb search -l dotnet --text "dispose pattern"

Available Flags

FlagTypeDefaultDescription
--textstring—Natural language search query (mutually exclusive with --path)
--pathstring—Path to a file containing query text or code (mutually exclusive with --text)
--idslist—One or more knowledge base item IDs to retrieve directly (no other flags allowed)
--language / -lstring—Language to search (required for all non---ids queries)
--markdownflagoffRender output as markdown instead of JSON (not allowed with --ids)
Show full SKILL.md (258 more words)Show less

Understanding the Output

Markdown Format (--markdown)

Each result is rendered as a section with:

  • Metadata block: {kind}_id (e.g. guideline_id or memory_id), score (if available), and exception status (if true)
  • Title: The guideline or memory title
  • Content: The full guideline or memory text
  • Examples: Good and bad code examples in fenced code blocks

This is the same format the AI reviewer sees during RAG-based reviews.

JSON Format (default)

Returns the full Context object with nested guidelines, examples, and memories including scores and all linked entities. Useful for programmatic inspection.

ID Lookup Format

Returns SearchItem objects as a JSON array with a curated projection of index fields: id, kind, title, content, language, service, is_exception, example_type, and search scores.

Gotchas

  • --ids is exclusive: When using --ids, do not pass --language, --text, --markdown, or --path. The command will error.
  • --language is required for all non---ids queries: Both --text and --path require --language to filter the index.
  • Exactly one of --text or --path: For non-ID queries, provide one but not both.
  • Use python cli.py not avc/avc.bat: The wrapper scripts (avc on Linux/macOS, avc.bat on Windows) invoke the system python/python3, which may not be the correct virtual environment. Run python cli.py directly to ensure the right interpreter is used.
  • Do NOT use 2>&1: This merges stderr log messages into the output, corrupting it.
  • Prefer --markdown for text/path queries: The markdown output is more readable and is what the AI reviewer actually sees. Use JSON only when you need structured data.
  • Scores are normalized: Search result scores use Z-score normalization (mean=50). Higher is more relevant.

© Azure, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

Just SKILL.md in packages/python-packages/apiview-copilot/.github/skills/kb-search of Azure/azure-sdk-tools.

Open the folder on GitHubat commit cc5ca24

Compare with similar skills

Kb Search next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.

Kb Search compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Kb Search this skillAzure/azure-sdk-tools134—~1.6kAutomated safety check: PassMIT
Foundry Iqmicrosoft/GitHub-Copilot-for-Azure255—~872Automated safety check: PassMIT
Datachain Knowledgedatachain-ai/datachain2.8k—~3kAutomated safety check: PassApache-2.0
Azure Search Documents TSmicrosoft/skills3.1k—~1.8kAutomated safety check: PassMIT
Microsoft Skill CreatorMicrosoftDocs/mcp1.9k3 repos~2.1kAutomated safety check: PassCC-BY-4.0
Dev ServerAzure/cosmos-explorer131—~1.6kAutomated safety check: PassMIT

Similar skills

  • Foundry Iq

    microsoft/GitHub-Copilot-for-Azure

    Official

    Foundry IQ knowledge bases. An agent skill from microsoft/GitHub-Copilot-for-Azure.

    255 GitHub stars~872 tokensUpdated yesterday
    Knowledge ManagementAuto-check passed

  • Datachain Knowledge

    datachain-ai/datachain

    A skill your agent uses whenever datasets, cloud storage buckets, or data pipelines are mentioned — creating, saving, querying, listing, exploring, deleting, or processing data in S3, GCS, Azure…

    2.8k GitHub stars~3k tokensUpdated today
    Knowledge ManagementAuto-check passed

  • Azure Search Documents TS

    microsoft/skills

    Official

    Build search applications using Azure AI Search SDK for JavaScript (@azure/search-documents).

    3.1k GitHub stars~1.8k tokensUpdated today
    Knowledge ManagementAuto-check passed

  • Microsoft Skill Creator

    MicrosoftDocs/mcp

    Official

    Create agent skills for Microsoft technologies using official documentation.

    1.9k GitHub starsUsed in 3 repos~2.1k tokens
    Agent WorkflowsAuto-check passed

  • Dev Server

    Azure/cosmos-explorer

    Official

    Start the local webpack dev server and connect to it with the Playwright browser.

    131 GitHub stars~1.6k tokensUpdated 2 days ago
    Testing & QAAuto-check passed

  • Azure Storage

    microsoft/GitHub-Copilot-for-Azure

    Official

    Azure Storage Services including Blob Storage, File Shares, Queue Storage, Table Storage, and Data Lake.

    255 GitHub starsUsed in 2 repos~1.3k tokens
    DatabasesAuto-check passed

More from Azure/azure-sdk-tools

All 35 skills in this repo

  • Apiview Feedback Resolution

    Azure/azure-sdk-tools

    Official

    Analyze and resolve APIView review feedback on Azure SDK PRs.

    134 GitHub stars~547 tokensUpdated today
    Auto-check passed

  • Azsdk Common Live And Recorded Tests

    Azure/azure-sdk-tools

    Official

    Deploy test resources and run Azure SDK tests in live, record, or playback mode.

    134 GitHub stars~1.5k tokensUpdated today
    Auto-check: notes

  • Azsdk Common Pipeline Analysis

    Azure/azure-sdk-tools

    Official

    Analyze Azure SDK CI/CD pipeline failures into a structured diagnosis, and define the required output format.

    134 GitHub stars~1.2k tokensUpdated today
    Auto-check passed

  • Azsdk Common Prepare Release Plan

    Azure/azure-sdk-tools

    Official

    Create, get, update, abandon, and link SDK PRs to release plan work items for Azure SDK releases.

    134 GitHub stars~2.7k tokensUpdated today
    Auto-check passed

  • Azure Typespec Assessment

    Azure/azure-sdk-tools

    Official

    Assess Azure TypeSpec Git diffs for semantic intent, REST and downstream SDK breaking changes, Azure Guidelines compliance, and documentation completeness.

    134 GitHub stars~1.9k tokensUpdated today
    Auto-check passed

Works with

Categories

Questions about Kb Search

What does Kb Search do?

Query the APIView Copilot knowledge base for guidelines, examples, and memories. Kb Search is an agent skill from Azure/azure-sdk-tools, published by the product's own GitHub organization. Query the APIView Copilot knowledge base for guidelines, examples, and memories.

When should I use Kb Search?

Kb Search fits situations like: search knowledge base; lookup guideline; search guidelines; search examples.

How do I install Kb Search in Claude Code?

Run `npx skills add Azure/azure-sdk-tools --skill kb-search -a claude-code`. Or copy the skill folder (packages/python-packages/apiview-copilot/.github/skills/kb-search in Azure/azure-sdk-tools) into .claude/skills/kb-search in your project. Claude Code loads it when a task matches its description.

How do I install Kb Search in Codex?

Run `npx skills add Azure/azure-sdk-tools --skill kb-search -a codex`. Or copy the skill folder (packages/python-packages/apiview-copilot/.github/skills/kb-search in Azure/azure-sdk-tools) into .agents/skills/kb-search in your project. Codex loads it when a task matches its description.

Can I use Kb Search in Cursor, Gemini CLI or GitHub Copilot?

Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add Azure/azure-sdk-tools --skill kb-search -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/kb-search, .gemini/skills/kb-search, .github/skills/kb-search and .opencode/skills/kb-search in your project.

What does Kb Search need to run?

Going by SKILL.md and its folder, Kb Search needs the command-line tools its instructions call (python). Our summary lists: Python 3.

Does Kb Search access the network?

SKILL.md contains no URLs. Any network use would come from the scripts or tools the agent runs. This is read from the text; nothing was executed.

Is Kb Search safe to install?

Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.

What licence does Kb Search use?

Kb Search is published under the MIT licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Kb Search use?

About 1.6k tokens (SKILL.md is roughly 6.5k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full.

What are the alternatives to Kb Search?

Skills that share tags, products or a category with Kb Search: Foundry Iq (microsoft/GitHub-Copilot-for-Azure, 255 stars), Datachain Knowledge (datachain-ai/datachain, 2.8k stars), Azure Search Documents TS (microsoft/skills, 3.1k stars) and Microsoft Skill Creator (MicrosoftDocs/mcp, 1.9k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Kb Search?

Azure (a GitHub organization, an official publisher) maintains it in Azure/azure-sdk-tools, which has 134 GitHub stars. The repository holds 35 skills in this directory. The repository was last updated on October 9, 2026.

Source: Azure/azure-sdk-tools on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.