launchdarkly-flag-command

Resolve `/flag` style requests into the right LaunchDarkly flag lookup flow. Use when the user types `/flag`, asks to quickly find a flag by name/key, wants a direct flag detail summary, or needs fast disambiguation between similar flags.

25stars8forksUpdated 8/13/2026

MCP Tools DevOps & CI/CD #devops #feature-flags #mcp #read-only #launchdarkly

Security Assessment

Safe(95/100)

Security Score95/100

About launchdarkly-flag-command

LaunchDarkly Flag Command is a read-only lookup entrypoint that resolves quick `/flag` style requests into the right LaunchDarkly flag flow. It parses user intent, resolves a flag from a fuzzy query with minimal friction, returns an actionable summary, and routes forward to deeper workflows when the request calls for more. It solves the problem of fast flag disambiguation and detail retrieval without pulling the user into a heavier lifecycle workflow prematurely.

The skill is deliberately constrained. It defines a hard scope boundary: it must not create, toggle, update, or delete flags, and must not assess whether a flag is safe to remove, stale, or ready for cleanup — questions about removal or staleness are answered only with the flag summary plus a fixed routing message pointing to the dedicated flag discovery or flag cleanup skills. The workflow parses and normalizes input (query text, optional environment, project, and search-preference hints), resolves the flag with the `list-flags` MCP tool and then `get-flag` for details, and returns a concise summary (key and name, environment state, off variation and fallthrough, rule complexity, and a direct URL). Optional MCP tools support cross-environment status and health snapshots. It requires the remotely hosted LaunchDarkly MCP server to be configured.

The target users are engineers and platform teams using LaunchDarkly who want a fast, safe flag lookup command inside their agent. Use cases include finding a flag by fuzzy name, disambiguating between similar flags, getting a quick config summary before acting, and being routed to the correct create, targeting, discovery, or cleanup skill for deeper work.

FAQ

What can this skill do, and what can't it do?

It performs read-only flag lookups and disambiguation and returns a config summary. By hard constraint it will not create, toggle, update, or delete flags, and it will not judge whether a flag is safe to remove or stale.

What are the prerequisites?

The remotely hosted LaunchDarkly MCP server must be configured. It requires the list-flags and get-flag tools, and can optionally use get-flag-status-across-envs and get-flag-health.

What happens if I ask whether a flag is safe to remove?

It shows only the flag summary and then a fixed routing message directing you to the flag discovery or flag cleanup skills, which scan code references and cross-environment status. It gives no verdict or removal steps itself.

How does it resolve which flag I mean?

It uses list-flags to search your query; on a single clear match it resolves directly, on multiple matches it returns a short disambiguation list (key, name, state) for you to pick, and on no match it suggests one broader query.

What default environment does it use?

It defaults to the production environment unless you specify another, such as staging, and treats project and environment as first-class context rather than making hidden assumptions.