Add Agent Builder skills + dashboards/viz documentation#5868
Merged
Conversation
Introduces three new pages for the Agent Builder skills feature: - `skills.md`: Overview of how skills work, when to use them, how to create custom skills via UI and API, and how to assign them to agents. - `skill-creation-guidelines.md`: Structured guidance for writing effective skill descriptions and instructions, with annotated examples for trigger conditions, core instructions, edge cases, and referenced content. Includes a decision guide for choosing between skills, the system prompt, and direct conversation. - `builtin-skills-reference.md`: Full reference listing all built-in skills with applies_to tags, anchor IDs, and an inline tools section. Also updates existing pages to integrate skills into the broader Agent Builder docs: Skills APIs added to `kibana-api.md`, cross-references added to `prompt-engineering.md`, `tools.md`, `builtin-tools-reference.md`, and `custom-agents.md`. All three new pages wired into `toc.yml`.
Contributor
Contributor
Vale Linting ResultsSummary: 2 warnings, 4 suggestions found
|
| File | Line | Rule | Message |
|---|---|---|---|
| explore-analyze/ai-features/agent-builder/builtin-skills-reference.md | 88 | Elastic.Spelling | 'chatbot' is a possible misspelling. |
| explore-analyze/ai-features/agent-builder/tools.md | 17 | Elastic.EndPuntuaction | Don't end headings with punctuation. |
💡 Suggestions (4)
| File | Line | Rule | Message |
|---|---|---|---|
| explore-analyze/ai-features/agent-builder/builtin-skills-reference.md | 51 | Elastic.Semicolons | Use semicolons judiciously. |
| explore-analyze/ai-features/agent-builder/builtin-skills-reference.md | 51 | Elastic.Semicolons | Use semicolons judiciously. |
| explore-analyze/ai-features/agent-builder/builtin-skills-reference.md | 65 | Elastic.WordChoice | Consider using 'atypical, not typical' instead of 'abnormal', unless the term is in the UI. |
| explore-analyze/ai-features/agent-builder/skills.md | 17 | Elastic.WordChoice | Consider using 'efficient' instead of 'easy', unless the term is in the UI. |
The Vale linter checks documentation changes against the Elastic Docs style guide.
To use Vale locally or report issues, refer to Elastic style guide for Vale.
KDKHD
reviewed
Apr 13, 2026
Co-authored-by: Kenneth Kreindler <42113355+KDKHD@users.noreply.github.com>
…lt-in skills Co-authored-by: Kenneth Kreindler <42113355+KDKHD@users.noreply.github.com>
Co-authored-by: Kenneth Kreindler <42113355+KDKHD@users.noreply.github.com>
Co-authored-by: Kenneth Kreindler <42113355+KDKHD@users.noreply.github.com>
Co-authored-by: Kenneth Kreindler <42113355+KDKHD@users.noreply.github.com>
- Rename headings to consistent imperative form throughout
- Promote single-responsibility principle to its own H2 ("Scope each skill to one task")
- Promote semantically-distinct descriptions to H3 with checklist entry
- Expand checklist with sub-headings for all sections
- Add "Test your skill" section; remove defensive disclaimer from intro
- Fix misleading callout on "Too vague" description example
- Surface buried guidance (vague openings, tool documentation) via tip admonitions and paragraph breaks
- Replace removed alert-analysis skill example in skills.md with entity-analytics - Drop #inline-tools fragment from tools.md and builtin-tools-reference.md (no standalone anchor exists)
neptunian
reviewed
Apr 13, 2026
neptunian
reviewed
Apr 13, 2026
…e example; add alert investigation mention Agent-Logs-Url: https://github.com/elastic/docs-content/sessions/228c17be-926d-4046-ae10-c7da3c0dae84 Co-authored-by: leemthompo <32779855+leemthompo@users.noreply.github.com>
neptunian
approved these changes
Apr 13, 2026
KDKHD
approved these changes
Apr 14, 2026
## Summary - Add new `dashboards-and-visualizations.md` page documenting how agents create, edit, and save dashboards and Lens visualizations through chat - Uncomment dashboard tools (`dashboard.create_dashboard`, `dashboard.update_dashboard`) and `platform.core.create_visualization` in built-in tools reference - Add cross-reference from core dashboard docs (`create-dashboard.md`) - Add brief mention + link in `chat.md` pointing to the new page - Add new page to `toc.yml` as a child of `chat.md` Closes elastic/docs-content-internal#1008 **Note:** Screenshot placeholders are wired up in the new page. Image files still need to be added to `explore-analyze/ai-features/agent-builder/images/`: 1. `agent-builder-dashboard-canvas-preview.png` 2. `agent-builder-dashboard-save-dialog.png` 3. `agent-builder-dashboard-outdated-notification.png` ## TODO - [ ] Verify new page renders correctly in docs-builder - [ ] Verify toc.yml entry places the page under Chat > Dashboards and visualizations - [ ] Verify cross-links resolve (skills reference anchors, tools reference anchors, create-dashboard.md link) - [ ] Add screenshot image files and verify they render - [ ] Coordinate merge with #5850 (Charlotte's UI screenshot updates touch some of the same files) 🤖 Generated with [Claude Code](https://claude.com/claude-code) ## Preview (click on me) <img width="2896" height="10722" alt="Screenshot 2026-04-16 at 14-44-41 Dashboards and visualizations in Elastic Agent Builder chat Elastic Docs" src="https://github.com/user-attachments/assets/a51db003-fca8-431d-ac44-c856df8d79df" /> --------- Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>
- skills.md: Lead opening with user value before the definition - custom-skills.md: Add "When to create" section, outcome sentence, test step in stepper, and tip for import vs. create inline - skill-creation-guidelines.md: Use action verb title, add Next steps, deduplicate Related pages Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
benironside
approved these changes
Apr 17, 2026
Contributor
benironside
left a comment
There was a problem hiding this comment.
I reviewed the new pages only. They look great. Left one small suggestion and a couple minor edits. Everything LGTM. Approving. The Skill creation guidelines in particular look excellent. Good job!
mdbirnstiehl
approved these changes
Apr 17, 2026
Member
mdbirnstiehl
left a comment
There was a problem hiding this comment.
Nice work, @leemthompo! 🥇
|
|
||
| # Dashboards and visualizations in {{agent-builder}} chat | ||
|
|
||
| Agents can create and manage [dashboards](/explore-analyze/dashboards.md) and [visualizations](/explore-analyze/visualize.md) directly in the conversation. Ask a question like "create a dashboard to monitor host CPU and memory usage" and the agent builds a dashboard with {{esql}}-powered visualization panels. |
Member
There was a problem hiding this comment.
A nit and not sure how we want to talk about dialog with the agent, but not really a question.
leemthompo
commented
Apr 19, 2026
Co-authored-by: Benjamin Ironside Goldstein <91905639+benironside@users.noreply.github.com> Co-authored-by: Liam Thompson <leemthompo@gmail.com>
leemthompo
commented
Apr 19, 2026
Co-authored-by: Liam Thompson <leemthompo@gmail.com>
charlotte-hoblik
approved these changes
Apr 20, 2026
3 tasks
leemthompo
added a commit
that referenced
this pull request
Apr 22, 2026
…hods (#6059) ## [URL preview](https://docs-v3-preview.elastic.dev/elastic/docs-content/pull/6059/explore-analyze/ai-features/agent-builder/skills) ## Summary - Adds **Skills** as a fourth key concept on the Agent Builder overview page alongside Agent Chat, Agents, and Tools - Enriches the `skills.md` intro with a structured breakdown of what a skill bundles (instructions, tools, context) and how skills differ from tools and system prompts - Documents the three skill invocation methods: automatic discovery, slash commands (with screenshot), and attachment-driven activation - Updates `get-started.md` final step to lead with skills as the primary extension point - Links use case examples to the Security and Observability solution-specific Agent Builder pages ## Related - Tracks against #5719 (core AB collab items) - Builds on #5868 (skills docs: concept page, custom skills, built-in skills reference, skill creation guidelines) - Coordinates with #5850 (9.4 UI refresh, which adds skills mentions to the agents overview page) ## Test plan - [x] Verify cross-links resolve: `skills.md`, `builtin-skills-reference.md`, `custom-skills.md`, `skill-creation-guidelines.md`, solution-specific pages - [x] Verify `skill-slash-command.png` renders at correct size - [x] Confirm no merge conflicts with #5850 on `get-started.md` 🤖 Generated with [Claude Code](https://claude.com/claude-code) --------- Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
benironside
added a commit
that referenced
this pull request
Apr 29, 2026
## Summary Fixes #5718, #5719, and #5720. ### New pages - **`solutions/security/ai/agent-builder/skills-model.md`** — "Agents, skills, and tools" conceptual page for Security. Explains the one-agent-many-skills model in 9.4, with a comparison table (Agent / Skill / Tool), an Enable/manage section, the Threat Hunting Agent → skill relationship, and a short FAQ - **`solutions/security/ai/agent-builder/skills-use-cases.md`** — "Security use cases for Agent Builder." One section per SOC workflow, each naming the built-in skill(s) to enable and 3 example prompts grounded in what the skill actually does (pulled from the Kibana source) ### Updated pages - **`solutions/security/ai/agent-builder/agent-builder.md`** — added an `applies-switch` so 9.3 keeps the Threat Hunting agent content and 9.4 shows the skills-based content, with links to the two new pages. - **`explore-analyze/ai-features/agent-builder/builtin-agents-reference.md`** — Threat Hunting Agent section marked `deprecated 9.4`, with a prominent deprecation warning and a migration-path block pointing to the `threat-hunting` skill. - **`explore-analyze/ai-features/agent-builder/builtin-skills-reference.md`** — added two Security skills that were missing: - `alert-analysis` (GA 9.4+) - `automatic_troubleshooting` (preview 9.4, experimental feature flag) Also added "Supports ES|QL rule type only" to `detection-rule-edit` (real constraint from the source). - **`explore-analyze/ai-features/ai-chat-experiences/ai-agent-or-ai-assistant.md`** — applied your comparison-table updates for 9.4 (via `applies-switch` so 9.3 preserves the old tables): - **`solutions/toc.yml`** — adds the two new Security pages as children of `agent-builder.md`, concepts first then use cases. ## Attention @dhru42: Please double check the following: - [ ] **`alert-analysis` skill** — it's live in the Kibana source but wasn't in the #5868 built-in skills update. Confirm this should ship in 9.4 GA docs and that my description is right. - [ ] **`automatic_troubleshooting` skill** — source registers it behind the `automaticTroubleshootingSkill` experimental flag. Confirm preview + flag-gated framing is right, or whether it should be held out of docs entirely for 9.4. - [ ] **Example prompts** — grounded in Kibana source, but you may have better real-world prompts from demos/POCs. ## What's not here - **Release notes** — will be handled separately. - **Screenshots** — need a test env then I'll add some. ## Generative AI disclosure 1. Did you use a generative AI (GenAI) tool to assist in creating this contribution? - [ x] Yes Claude 4.6 & 4.7 / Cursor - [ ] No
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Contributes to https://github.com/elastic/docs-content-internal/issues/1041
closes https://github.com/elastic/docs-content-internal/issues/1008
Note
Dashboards component was approved in #5915
👀 URL preview
Summary
skills.md(overview),skill-creation-guidelines.md(guidelines), andbuiltin-skills-reference.md(reference)kibana-api.mdwith examples for all five CRUD operationsprompt-engineering.md,tools.md, andbuiltin-tools-reference.mdtoc.ymlNew pages
skills.md— Overview: what skills are, how agents use them, built-in vs custom, create via UI and API, add to agent, Skills API list.skill-creation-guidelines.md— Structured guidance for writing effective skill descriptions and instructions. Covers deciding where to put instructions (skill vs system prompt), writing clear and semantically distinct descriptions, structuring instructions with trigger conditions and edge cases, scoping each skill to one task, usingreferenced_content, assigning only relevant tools, and testing the skill. Includes annotated code examples throughout.builtin-skills-reference.md— Full reference listing all located built-in skills with inlineapplies_totags, section-level serverless scoping,$$$anchor-ids$$$for deep linking, and an inline tools section.Open items
search.*skills (ES project): source files not yet located in Kibana — descriptions are from UI only, tools unconfirmeddashboard-management: confirm whether the Experimental flag applies in all projects or only Securityobservability.rca: confirm GA vs preview — markedexperimental: truein Kibana sourceobservability.investigation: commented out pending elastic/kibana#262293