Atlas / Skills / composiohq / Launch Darkly Automation

Launch Darkly AutomationSAFE

skills/composiohq/launch-darkly-automation

A curated list of awesome Claude Skills, resources, and tools for customizing Claude AI workflows

Verdict
SAFE
Grade
B
Trust score
89 /100
Version
—
Hosts
—
License
—
Stars
76,829
01

Overview

A curated list of awesome Claude Skills, resources, and tools for customizing Claude AI workflows

Read from source at commit 661fa6b4bb7bOBSERVED · 2026-10-11
02

What it tells the agent

The instruction file, verbatim from the audited commit — this is the text the model reads, and the surface the audit's instruction layer examines. Quoted here so you can judge it without cloning anything.

---
name: LaunchDarkly Automation
description: "Automate LaunchDarkly feature flag management -- list projects and environments, create and delete trigger workflows, and track code references via the Composio MCP integration."
requires:
  mcp:
    - rube
---

# LaunchDarkly Automation

Automate your LaunchDarkly feature flag workflows -- enumerate projects and environments, create webhook-driven flag triggers, manage trigger lifecycle, and audit code references across repositories.

**Toolkit docs:** [composio.dev/toolkits/launch_darkly](https://composio.dev/toolkits/launch_darkly)

---

## Setup

1. Add the Composio MCP server to your client: `https://rube.app/mcp`
2. Connect your LaunchDarkly account when prompted (API key authentication)
3. Start using the workflows below

---

## Core Workflows

### 1. List Projects

Use `LAUNCH_DARKLY_LIST_PROJECTS` to discover all projects and their keys for subsequent operations.

```
Tool: LAUNCH_DARKLY_LIST_PROJECTS
Inputs:
  - filter: string (e.g., "query:myproject" or "keys:proj1,proj2" or "tags:mytag")
  - expand: string (e.g., "environments" to include env list per project)
  - limit: integer (default 20)
  - offset: integer (pagination start index)
  - sort: string (e.g., "name" or "-name" for descending)
```

### 2. Get Environments for a Project

Use `LAUNCH_DARKLY_GET_ENVIRONMENTS` to list all environments within a project (production, staging, test, etc.).

```
Tool: LAUNCH_DARKLY_GET_ENVIRONMENTS
Inputs:
  - project_key: string (required) -- e.g., "my-project", "default"
  - filter: string (e.g., "query:production")
  - limit: integer (default 20)
  - offset: integer (pagination)
  - sort: string (e.g., "name" or "-name")
```

### 3. Create a Flag Trigger Workflow

Use `LAUNCH_DARKLY_CREATE_TRIGGER_WORKFLOW` to set up automated flag toggles triggered by external events (webhooks, Datadog alerts, etc.).

```
Tool: LAUNCH_DARKLY_CREATE_TRIGGER_WORKFLOW
Inputs:
  - project_key: string (required)
  - feature_flag_key: string (required) -- e.g., "new-feature", "enable-dark-mode"
  - environment_key: string (required) -- e.g., "production", "staging"
  - integration_key: string (default "generic-trigger") -- or "datadog", "honeycomb", "dynatrace"
  - instructions: array of objects (optional):
      - kind: "flag_action" (fixed)
      - action: "turnFlagOn" | "turnFlagOff"
  - comment: string (optional) -- description of the trigger purpose
```

The trigger generates a unique webhook URL that can be called to execute the configured flag action.

### 4. Delete a Flag Trigger Workflow

Use `LAUNCH_DARKLY_DELETE_TRIGGER_WORKFLOW` to permanently remove a trigger and its URL.

```
Tool: LAUNCH_DARKLY_DELETE_TRIGGER_WORKFLOW
Inputs:
  - project_key: string (required)
  - feature_flag_key: string (required)
  - environment_key: string (required)
  - id: string (required) -- the trigger ID returned during creation
```

**Warning:** Deletion is irreversible. The trigger and its URL cannot be recovered.

### 5. List Code Reference Repositories

Use `LAUNCH_DARKLY_LIST_CODE_REFERENCE_REPOSITORIES` to track where feature flags are used in your codebase.

```
Tool: LAUNCH_DARKLY_LIST_CODE_REFERENCE_REPOSITORIES
Inputs:
  - projKey: string (optional) -- filter by project key
  - flagKey: string (optional) -- filter by feature flag key
  - withBranches: string (any value to include branch data)
  - withReferencesForDefaultBranch: string (any value to include code refs for default branch)
```

**Note:** Code references is an Enterprise feature requiring `code-reference-repository` write permissions.

---

## Known Pitfalls

| Pitfall | Detail |
|---------|--------|
| Project key discovery | Always use `LAUNCH_DARKLY_LIST_PROJECTS` first to find valid project keys before calling other tools. |
| Environment key format | Environment keys are lowercase slugs (e.g., "production", "test"), not display names. |
| Trigger deletion is permanent | Once deleted via `LAUNCH_DARKLY_DELETE_TRIGGER_WORKFLOW`, the trigger URL is unrecoverable. |
| Enterprise-only code refs | `LAUNCH_DARKLY_LIST_CODE_REFERENCE_REPOSITORIES` requires Enterprise plan and write permissions. |
| Trigger instructions format | Each instruction object requires `kind: "flag_action"` (fixed constant) and `action` as either `turnFlagOn` or `turnFlagOff`. |

---

## Quick Reference

| Tool Slug | Description |
|-----------|-------------|
| `LAUNCH_DARKLY_LIST_PROJECTS` | List all projects with filtering and pagination |
| `LAUNCH_DARKLY_GET_ENVIRONMENTS` | List environments within a project |
| `LAUNCH_DARKLY_CREATE_TRIGGER_WORKFLOW` | Create a webhook-driven flag trigger |
| `LAUNCH_DARKLY_DELETE_TRIGGER_WORKFLOW` | Permanently delete a flag trigger |
| `LAUNCH_DARKLY_LIST_CODE_REFERENCE_REPOSITORIES` | List repos with code references to flags |

---

*Powered by [Composio](https://composio.dev)*
03

Trust audit

SAFEgrade B · trust 89/100 Nothing in the source contradicts what it says it does. Grade A is reserved for packages that have also passed the behavioural sandbox.

LayerWhat it checksResult
L0Provenance & inventoryPASS
L1Static analysis of the codeNA
L2Instruction surface (what it tells the agent)PASS
L3Class-specific surfacePASS
L4Behavioural (sandbox)SKIPPED

What the source does

Filesystem
none-observed
Network
none-observed
Shell
none-observed
Dependencies
pinned
Secrets in source
none-found

Findings (0)

No findings outside the package's declared scope.

Gates applied: no_behavioural_pass.

Audited 2026-10-11 · audit v0.4.1 · source sha 661fa6b4bb7bfull audit observations/trust-audit/skill/composiohq__launch-darkly-automation.json · Report an issue / request a re-scan
04

Audit history

Every audit this skill has had.

DateSourceVerdictGradeScoreChange
2026-10-11661fa6b4bb7bSAFEB89first audit
05

Questions

What does the Launch Darkly Automation skill do?

A curated list of awesome Claude Skills, resources, and tools for customizing Claude AI workflows

Is Launch Darkly Automation safe to install?

The audit found nothing in the source that contradicts what it says it does, and graded it B (89/100). Grade A is held back for packages that have also passed a sandboxed behavioural run, which is why a clean skill reads B.

What can Launch Darkly Automation access on my machine?

The audit observed no filesystem, network or shell use at all in its source.

How current is this page?

The grade is for one exact copy of the source (661fa6b4bb7b), read on 2026-10-11. The repository is watched, and a new audit runs when it changes — this is the first audit.

Advertisement