Docs fix — frontmatter: path /explore-analyze/ai-features — 20 pages

Generated by `gh-aw-docs-frontmatter-sweep` for `elastic/docs-content` on 2026-W25.

Path /explore-analyze/ai-features · 75 total in scope · subtree corpus 75 pages.

## Findings (20)

```yaml
- file: explore-analyze/ai-features/ai-chat-experiences.md
line: 1
category: missing-description
severity: high
evidence: "frontmatter has no `description` field"
suggested_fix: |
description: "Overview of Kibana's AI chat options: Agent Builder for custom agents with specialized tools, and AI Assistant for knowledge base and data analysis."

- file: explore-analyze/ai-features/ai-chat-experiences/ai-assistant.md
line: 1
category: missing-description
severity: high
evidence: "frontmatter has no `description` field"
suggested_fix: |
description: "AI assistants in Kibana provide natural language interfaces for Elasticsearch search, Observability monitoring, and Security threat analysis."

- file: explore-analyze/ai-features/elastic-agent-builder.md
line: 1
category: missing-description
severity: high
evidence: "frontmatter has no `description` field"
suggested_fix: |
description: "Elastic Agent Builder is an AI platform for creating agents that answer questions and take actions over Elasticsearch data using natural language."

- file: explore-analyze/ai-features/llm-guides/connect-to-amazon-bedrock.md
line: 1
category: missing-description
severity: high
evidence: "frontmatter has no `description` field"
suggested_fix: |
description: "Step-by-step guide for setting up an Amazon Bedrock connector in Kibana to use Claude and other supported LLMs with Elastic AI features."

- file: explore-analyze/ai-features/llm-guides/connect-to-azure-openai.md
line: 1
category: missing-description
severity: high
evidence: "frontmatter has no `description` field"
suggested_fix: |
description: "Step-by-step guide for setting up an Azure OpenAI connector in Kibana to use Azure-hosted OpenAI models with Elastic AI features."

- file: explore-analyze/ai-features/llm-guides/connect-to-google-vertex.md
line: 1
category: missing-description
severity: high
evidence: "frontmatter has no `description` field"
suggested_fix: |
description: "Step-by-step guide for setting up a Google Vertex AI connector in Kibana to use Vertex AI LLMs with Elastic Security AI features."

- file: explore-analyze/ai-features/llm-guides/connect-to-lmstudio-observability.md
line: 1
category: missing-description
severity: high
evidence: "frontmatter has no `description` field"
suggested_fix: |
description: "Guide for connecting a local LLM to the Elastic Observability AI Assistant using LM Studio as the local model server."

- file: explore-analyze/ai-features/llm-guides/connect-to-lmstudio-security.md
line: 1
category: missing-description
severity: high
evidence: "frontmatter has no `description` field"
suggested_fix: |
description: "Guide for connecting a local LLM to Elastic Security AI features using LM Studio and an Nginx reverse proxy."

- file: explore-analyze/ai-features/llm-guides/connect-to-openai.md
line: 1
category: missing-description
severity: high
evidence: "frontmatter has no `description` field"
suggested_fix: |
description: "Step-by-step guide for setting up an OpenAI connector in Kibana to use OpenAI LLMs with Elastic AI features."

- file: explore-analyze/ai-features/llm-guides/connect-to-vLLM.md
line: 1
category: missing-description
severity: high
evidence: "frontmatter has no `description` field"
suggested_fix: |
description: "Guide for running an OpenAI-compatible LLM with vLLM behind an Nginx reverse proxy to connect Elastic AI features in air-gapped environments."

- file: explore-analyze/ai-features/llm-guides/llm-connectors.md
line: 1
category: missing-description
severity: high
evidence: "frontmatter has no `description` field"
suggested_fix: |
description: "Configure access to Elastic Managed LLMs or connect third-party LLMs via Amazon Bedrock, Azure OpenAI, Google Vertex, and other supported connectors."

- file: explore-analyze/ai-features/llm-guides/local-llms-overview.md
line: 1
category: missing-description
severity: high
evidence: "frontmatter has no `description` field"
suggested_fix: |
description: "Overview of connector options for self-managed LLMs, for Elastic Security and Observability in air-gapped, restricted, or custom-model environments."

- file: explore-analyze/ai-features/manage-access-to-ai-assistant.md
line: 1
category: missing-description
severity: high
evidence: "frontmatter has no `description` field"
suggested_fix: |
description: "Control available LLM models, track token usage, install documentation, and configure settings for Kibana AI features from GenAI Settings."

- file: explore-analyze/ai-features/agent-builder/a2a-server.md
line: 3
category: weak-description
severity: medium
evidence: "description opens with teaching phrase 'Learn how to'"
suggested_fix: |
description: "Interact with Agent Builder agents from external clients using A2A protocol endpoints and API key authentication."

- file: explore-analyze/ai-features/agent-builder/agent-builder-agents.md
line: 3
category: weak-description
severity: medium
evidence: "description opens with teaching phrase 'Learn how'"
suggested_fix: |
description: "Overview of how Agent Builder agents use tools to solve problems, with a comparison of built-in and custom agents and the iterative reasoning loop."

- file: explore-analyze/ai-features/agent-builder/agent-builder-api-tutorial.md
line: 3
category: weak-description
severity: medium
evidence: "description opens with teaching phrase 'Learn how to'"
suggested_fix: |
description: "Step-by-step tutorial for creating custom agents and tools using the Agent Builder APIs, working with sample book data."

- file: explore-analyze/ai-features/agent-builder/agents-and-workflows.md
line: 3
category: weak-description
severity: medium
evidence: "description opens with teaching phrase 'Learn how to'"
suggested_fix: |
description: "Invoke AI agents in Elastic Workflows using the ai.agent or kibana.request steps to add reasoning and decision-making to automated tasks."

- file: explore-analyze/ai-features/agent-builder/builtin-skills-reference.md
line: 3
category: weak-description
severity: medium
evidence: "description is a terse label with no indication of what the skills do or why they matter"
suggested_fix: |
description: "Reference for all built-in Agent Builder skills, which bundle domain-specific instructions and tools that agents activate for specialized tasks."

- file: explore-analyze/ai-features/agent-builder/chat.md
line: 3
category: weak-description
severity: medium
evidence: "description opens with teaching phrase 'Learn how to'"
suggested_fix: |
description: "Chat with AI agents in Agent Builder, inspect reasoning steps, and manage conversation history through the Agent Builder UI or API."

- file: explore-analyze/ai-features/agent-builder/connectors.md
line: 3
category: weak-description
severity: medium
evidence: "description opens with teaching phrase 'Learn about'"
suggested_fix: |
description: "The Agent Builder connectors library configures agent access to external systems and data sources beyond Elasticsearch."
```

## Done when
- All listed pages have a non-empty, <=200-character `description` field and required frontmatter keys present per the repo's frontmatter schema.
- A PR addressing this issue is merged.

## Notes
- Snippet files under `_snippets/` are present in the in-scope list but carry no frontmatter; they are intentionally excluded from this audit.
- +23 additional pages with weak descriptions will surface in the next sweep (including `custom-agents.md`, `custom-skills.md`, `get-started.md`, `mcp-server.md`, `models.md`, `permissions.md`, `skills.md`, `tools.md`, `tools/builtin-tools-reference.md`, `tools/custom-tools.md`, troubleshooting pages, and others).

> Generated by [Docs frontmatter sweep agent](https://github.com/elastic/docs-content/actions/runs/27846531185) · 296.7 AIC · ⌖ 13.7 AIC · ⊞ 27K · [◷](https://github.com/search?q=repo%3Aelastic%2Fdocs-content+is%3Aissue+%22gh-aw-workflow-call-id%3A+elastic%2Fdocs-content%2Fgh-aw-docs-frontmatter-sweep%22&type=issues)

Start with the listed pages under explore-analyze/ai-features, including ai-chat-experiences.md, the llm-guides files, and agent-builder files, then check the repository's frontmatter schema. Add or revise each listed description while preserving required keys; done means all 20 descriptions are non-empty, no longer than 200 characters, and a PR is merged.

Written by the indexing model from the issue text.