Index of every page on this site: llms.txt
All skills
launchdarkly avatar

/launchdarkly-experiment-guide

@f0ee110 official
Run it yourself

Guide a developer from 'I don't know what to test' to a ready-to-create LaunchDarkly experiment. Use when someone wants to run an experiment or A/B test but doesn't have a metric, a flag, or a hypothesis yet, or isn't sure what to measure. Scans the codebase and the connected LaunchDarkly account, proposes a hypothesis and metric from real context, wires up the flag and tracking code, then drafts the experiment and offers to create it. Built for engineers without a data-science or product-analytics background.

Use this Skill: https://skilld.dev/gh/launchdarkly/ai-tooling/launchdarkly-experiment-guide

Nothing lands on disk. Nothing to clean up.

Fork this Skill

Edit a local copy. It keeps the author and licence.

referencesdiscovery.md

≈739 tokens on demand. Your agent reads this file only when SKILL.md points to it.

Discovery guide — inferring the product and recent features

The goal of Step 1–2 is a fast, accurate read of what this app is and what the team has been building, not a full audit. Spend a handful of tool calls, then confirm with the user.

Infer the product

Read these first — they give the most signal per byte:

File What it tells you
README*, docs/ index The product in the team's own words.
package.json / pyproject.toml / go.mod / Gemfile / pom.xml / *.csproj Name, description, and the dependency list (framework, payment, auth, analytics libs).
Top-level dirs app/, pages/, routes/, controllers/, cmd/, services/ → the major surfaces.
Entrypoint src/index.*, main.*, app.*, server.* → app type (web/API/CLI/mobile).

Framework → product-type hints:

  • next, react, vue, svelte, @angular/core → web frontend / full-stack app.
  • express, fastify, nestjs, flask, django, fastapi, gin, rails → backend/API.
  • react-native, expo, flutter → mobile.
  • stripe, braintree, shopify → commerce / revenue surface (strong experiment target).
  • segment, amplitude, posthog, mixpanel, launchdarkly → analytics already present (reusable events).

Name the major user-facing surfaces you can see — checkout, onboarding/signup, search, feed, pricing, settings. These are where experiments live.

Surface recent features

Recent work is the best experiment candidate — the team already cares, and the code is fresh.

  • git log --oneline -n 30 — scan commit subjects for feature words ("add", "new", "redesign", "revamp", "v2", feature names).
  • git log --since="8 weeks ago" --name-only --pretty=format:"%s" — which files changed for recent features; cluster by directory to spot the feature area.
  • Look for feature flags already in the diff (variation(, boolVariation, useFlags, ldClient) — a recently-added flag with no experiment is an ideal candidate.

Detect the LaunchDarkly SDK

Grep the manifest/imports to find which SDK the code uses:

  • launchdarkly-node-server-sdk, @launchdarkly/node-server-sdk → Node server
  • launchdarkly-js-client-sdk, launchdarkly-react-client-sdk → browser/React
  • launchdarkly-python-server-sdk → Python
  • launchdarkly-go-server-sdk / gopkg.in/launchdarkly/... → Go
  • none found → the user hasn't installed an SDK yet; the flag/metric delegate skills handle SDK setup.

Confirm before building

End discovery with one or two sentences and a yes/no:

"This looks like a Next.js e-commerce storefront. Recent commits are about a redesigned checkout and a recommendation widget, and you're already using the LaunchDarkly React SDK. Want to experiment on one of those?"

A wrong guess here is cheap to fix now and expensive to fix after you've drafted the wrong experiment.

Source: SKILL.md on GitHub

No rule matched.

skilld matched fixed text patterns in SKILL.md and file names. Patterns miss obfuscated code.

skilld run checks every file with the same patterns. It asks for approval before it loads a Skill with a behavior marked Needs approval.

No alerts15d3 checks · Risk SAFE
  • Gen Agent Trust Hub15d

    This skill is a guided assistant for setting up A/B tests and experiments using LaunchDarkly. It scans the local codebase and git history to suggest relevant experiments and interacts with LaunchDarkly via MCP or an API token to manage flags and metrics. The skill follows best practices for credential handling and user consent.

  • Socket15d

    No alerts

  • Snyk15d

    Risk: LOW · No issues

Signed by skilld at f0ee110. This ties the file your Agent reads to that commit on GitHub. It does not review the instructions.

Last checked against GitHub yesterday.

Activeupdated 2 weeks ago
metadata
{
  "author": "launchdarkly",
  "version": "0.1.0-experimental"
}
Other metadata
compatibility
Uses the LaunchDarkly MCP server when it is configured; otherwise falls back to the LaunchDarkly REST API (needs a LaunchDarkly API token). Reads the local codebase and git history.

README badge

README badge for launchdarkly/ai-tooling/launchdarkly-experiment-guide