K6 Docs
Grafana LabsOfficialPublished by Grafana Labs
- Skill
- Content generator
- Cloud and DevOps
- Apache-2.0
- Low risk
- Has scripts
AutoGineer summary
Written by AutoGineer from the skill’s own files, not by the author
The k6-docs skill provides a structured approach to managing documentation across the k6 ecosystem. It ensures that user-facing documentation, TypeScript type definitions, and release notes remain accurate and synchronized with the k6 codebase. The skill enforces strict quality standards, including mandatory validation of all code examples against the current k6 master branch and visual verification of rendered documentation using browser-based tools.
What it does
- Drafts and reviews user documentation for k6 features
- Generates and maintains TypeScript type definitions
- Drafts release notes and changelogs for new versions
- Validates code examples by executing them against k6@master
- Verifies documentation rendering and link integrity
- Applies consistent style conventions across all documentation
When to use it
- When updating or creating k6 API reference documentation
- When drafting release notes for a new k6 version
- When modifying TypeScript type definitions for k6
- When reviewing documentation changes for accuracy and style
- When verifying that code examples in documentation are functional
Advantages
- Ensures all documentation code examples are tested and functional
- Maintains consistency across multiple repositories
- Prevents broken links and rendering issues through automated checks
- Automates the validation of TypeScript types
- Provides a clear, repeatable workflow for documentation maintenance
Usage examples
- “Review the new documentation for the k6 browser module.”
- The agent will check the documentation against the style guide, verify all links, and run the included code examples against k6@master.
- “Generate the release notes for the latest k6 update.”
- The agent will draft the release notes, ensuring all mentioned features are documented and examples are validated.
- “Update the TypeScript types for the k6 HTTP module.”
- The agent will update the type definitions and run tests to ensure they are compatible with the current k6 implementation.
Install
Download the skill as a ZIP:
https://autogineer.com/agent-skills/grafana/k6-docs/downloadUnzip it into the cross-client skills folder of your project:
mkdir -p .agents/skills && unzip k6-docs.zip -d .agents/skills
Documentation by Grafana Labs
k6 Documentation
Document or review k6 features across three repositories: k6-DefinitelyTyped (TypeScript types), k6-docs (user documentation), and k6 (release notes).
Workflow
Pick the workflow that matches the user's intent:
Write documentation: follow references/workflows/write.md
Review documentation: follow references/workflows/review.md
Quick References
Critical Rules
Never push automatically - always ask first
Never chain commands with
&∨- run each command separately (prevents failures)Only document user-facing features - not internal implementation
Use
1.for every numbered list item (not1.,2.,3.)Test every code example against k6@master before committing
No
Co-Authored-By: Claudeor AI attribution in commits
Validating examples (always run before committing)
Every code example in k6 documentation and release notes must execute cleanly against k6@master. The minimum loop:
# 1. cd into the k6 repo (not k6-docs or k6-DefinitelyTyped)
cd ~/path/to/k6
# 2. Make sure you're on master at the latest commit (the contract is k6@master)
# First check for local work: if this prints anything, stop and ask the user before switching branches
git status --porcelain
git checkout master
git pull
# 3. Run each example as a separate command (no &&)
go run . run /path/to/script.jsIf the run fails: read the error, fix the example in the source (doc or release note), re-run. Do not commit a change whose example doesn't run clean against current master. For the full multi-example workflow with parallel subagents, see references/testing-workflow.md.
Files
- ATTRIBUTION.mdAdded by AutoGineer832 B
- LICENSELicence file added by AutoGineer11.1 KiB
- SKILL.md2.8 KiB
- references/agent-browser-reference.md3.0 KiB
- references/checklists/documentation.md4.1 KiB
- references/checklists/release-notes.md783 B
- references/checklists/typescript.md752 B
- references/repository-structure.md3.9 KiB
- references/subagent-prompts.md3.4 KiB
- references/templates/documentation.md3.1 KiB
- references/templates/release-notes.md854 B
- references/testing-workflow.md2.3 KiB
- references/troubleshooting.md4.7 KiB
- references/typescript-patterns.md4.2 KiB
- references/workflows/review.md2.7 KiB
- references/workflows/write.md5.4 KiB
- scripts/check_agent_browser.shScript524 B
- scripts/fetch_pr.shScript195 B
- scripts/find_branch.shScript968 B
- scripts/format_typescript.shScript236 B
- scripts/test_typescript.shScript303 B
- scripts/update_team_members.shScript1.4 KiB
- scripts/validate_links.shScript1.1 KiB
references/agent-browser-reference.md
agent-browser Command Reference
agent-browser is the PRIMARY tool for visual verification of documentation. Use it to catch rendering issues, broken styles, and visual problems that HTTP status checks cannot detect.
Installation
Check if installed:
which agent-browserIf not installed:
npm install -g [email protected]
agent-browser installNavigation
agent-browser open <url>- Navigate to URLagent-browser back- Go back in historyagent-browser reload- Reload current page
Inspection
agent-browser snapshot- Get accessibility tree with element @refsagent-browser snapshot -i- Interactive elements onlyagent-browser screenshot [path]- Screenshot (add--fullfor full page)agent-browser get text <selector>- Get element textagent-browser get title- Get page titleagent-browser get url- Get current URL
Interaction
agent-browser click <selector|@ref>- Click elementagent-browser type <selector> <text>- Type into elementagent-browser scroll <up|down|left|right> [pixels]- Scroll pageagent-browser wait <selector|milliseconds>- Wait for element or time
Selectors
Two types of selectors:
CSS selectors:
"a[href*='waitforrequest']"- Links containing text".docs-table td"- Table cells with class"h1"- Headers
@refs from snapshot:
Use
snapshotcommand first to get element @refsExample:
@e42references a specific elementMore reliable than CSS selectors
Sessions
--session <name>- Use isolated browser sessionDefault session persists between commands
Tips
Always use
--fullflag for full-page screenshots when verifying docsUse
snapshotto find elements before clicking (provides @refs)Combine with pipes:
agent-browser get text .title | grep "Error"to check for issuesCheck for 404s:
agent-browser get titleshould not return "404" or "Page Not Found"
Example Verification Workflow
# Step 1: Navigate to module index
agent-browser open "http://localhost:3002/docs/k6/next/javascript-api/k6-browser/page/"
# Step 2: Verify method appears in table
agent-browser screenshot --full index-page.png
# Step 3: Navigate to method docs
agent-browser open "http://localhost:3002/docs/k6/next/javascript-api/k6-browser/page/waitforevent/"
# Step 4: Verify rendering
agent-browser screenshot --full method-page.png
# Step 5: Check related links
agent-browser snapshot > snapshot.txt
agent-browser click "a[href*='waitforrequest']"
agent-browser screenshot related-link.png
# Step 6: Verify no 404
agent-browser get title # Should not be "404" or "Page Not Found"Fallback: curl
If agent-browser is unavailable, use curl as a last resort (provides no visual verification):
# Check HTTP status only
curl -s -o /dev/null -w '%{http_code}' http://localhost:3002/docs/k6/next/path/Note: curl only verifies pages return 200 - it cannot catch rendering, styling, or visual issues.
Requirements and compatibility
Works with: Claude Code, Claude apps, OpenAI Codex, GitHub Copilot, Cursor, Gemini CLI, Kiro, OpenCode, Goose, Amp, Roo Code, JetBrains Junie, Factory Droid, Devin (ex-Windsurf), Mistral Vibe, OpenHands, Letta
“Works with” means the tool reads the open Agent Skills format; a skill written for one tool may use features others lack.
Needs: Needs a command-line tool
Security
What our scanner noticed· Low risk
- Instructions that use the network
- Scripts that your agent may run
This skill includes scripts that your agent may run. Read them before you use it.
A static scan is not a guarantee. Skills are third-party content: review a skill before you let an agent use it.
Attribution and licence
- Published by
- Grafana Labs
- Publisher
- Grafana Labs (Official)
- Source
- grafana/skills · skills/grafana-k6/k6-docs · commit 8fb72a5
- Licence
- Apache License 2.0 (redistribution allowed)
- LICENSE (added by AutoGineer from LICENSE)
- ATTRIBUTION.md (written by AutoGineer)
- Synced
AutoGineer is not affiliated with or endorsed by Grafana Labs or Grafana Labs.
Download ZIP
The ZIP holds the skill’s files, its licence files and an ATTRIBUTION.md. Download k6-docs.zip