Name: launchdarkly-flag-create
Rating: 68 (1 reviews)
Author: launchdarkly

launchdarkly-flag-create

Create and configure LaunchDarkly feature flags in a way that fits the existing codebase. Use when the user wants to create a new flag, wrap code in a flag, add a feature toggle, or set up an experiment. Guides exploration of existing patterns before creating.

Quality

85%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Passed

No findings from the security scan

Quality

Content

75%Weight 40%Scale 1-5

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

A well-structured, efficient instruction-only skill: concrete MCP tool guidance, a clearly sequenced workflow ending in verification, and properly split reference files. The recurring gaps are the absence of an inline create-flag payload, no explicit fix-and-retry loop in the verify step, and cross-skill references to a sibling skill that may not be present in the environment.

Suggestions

Add a minimal inline `create-flag` payload example (key, kind, variations, temporary) in Step 3 so the core operation is executable from the body alone, not just via the flag-types reference.

Make the Step 5 verification a feedback loop — e.g., "If lint/build fails, fix the evaluation code and re-run; if `get-flag` shows wrong configuration, correct it before proceeding" — to reach explicit error recovery.

Guard the ../launchdarkly-flag-targeting/ references (e.g., "if the flag-targeting skill is available, see...; otherwise direct the user to the LaunchDarkly dashboard") so navigation does not break when the sibling skill is not installed.

Dimension	Reasoning	Score
Conciseness	The body is lean and assumes Claude's competence — no explanation of what feature flags or LaunchDarkly are — matching the "efficient; minor instances of over-explanation" anchor. It is not 5 because of minor repetition: the safe-default rule is stated in Step 4, Step 5, and "Important Context", and the opening paragraph restates the entire workflow that the Workflow section then details.	4 / 5
Actionability	Guidance is mostly executable: concrete search strings ("launchdarkly", "ldclient"), a decision table mapping intent to flag kind, copy-paste `update-flag-settings` payloads ("{kind: \"updateName\", value: \"New Name\"}"), and explicit defaults. It is not 5 because the central operation — the `create-flag` call — has no inline configuration example; the payload only exists in the referenced flag-types file, leaving a minor gap within the body itself.	4 / 5
Workflow Clarity	A clear five-step sequence (explore → determine type → create → add code → verify) with an explicit validation step ("Use `get-flag` to confirm", "Run the project's build or lint step"). It is not 5 because there is no explicit error-recovery feedback loop (e.g., what to do when lint fails or the flag config is wrong), and not 3 because validation checkpoints are present, not implicit.	4 / 5
Progressive Disclosure	The two in-bundle references (references/flag-types.md, references/sdk-evaluation-patterns.md) are real, one level deep, well signaled inline and in a References section, and the quick-decision table is appropriately kept inline. It is not 5 because the body repeatedly references ../launchdarkly-flag-targeting/SKILL.md and ../launchdarkly-flag-targeting/references/context-availability.md — paths outside this bundle that do not exist here — leaving navigation gaps for a user who has only this skill installed.	4 / 5
	Total	16 / 20 Passed

Description

90%Weight 40%Scale 1-5

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

A strong description: explicit "Use when..." triggers with natural synonyms, a clear third-person statement of what the skill does, and a distinct LaunchDarkly-scoped niche. The only weakness is that it lists just two concrete actions (create, configure), leaving the scope of what the skill actually walks through (exploration, evaluation-code wiring, verification) implicit.

Suggestions

Add one or two more concrete actions to the first sentence, e.g., "...including wiring evaluation code into the codebase and verifying the flag in LaunchDarkly", to lift specificity beyond two actions.

Optionally mention the exploration-first behavior as an action rather than a trailing clause ("Guides exploration of existing patterns before creating") so the what-coverage reads as capabilities, not process commentary.

Dimension	Reasoning	Score
Specificity	The description states two concrete actions — "Create and configure LaunchDarkly feature flags" — which matches the anchor for naming the domain plus 1-2 concrete actions. It is not level 4 because no further specific actions (e.g., wiring evaluation code, applying tags, setting defaults) are explicitly listed.	3 / 5
Completeness	It explicitly answers both what ("Create and configure LaunchDarkly feature flags in a way that fits the existing codebase") and when ("Use when the user wants to create a new flag, wrap code in a flag, add a feature toggle, or set up an experiment") with concrete trigger phrases — a clear anchor-5 match.	5 / 5
Trigger Term Quality	"create a new flag", "wrap code in a flag", "add a feature toggle", and "set up an experiment" are natural phrases users would say, covering flag/feature-toggle/experiment synonyms comprehensively for this domain. There are no relevant file extensions or additional common synonyms it omits, so it is not level 4.	5 / 5
Distinctiveness Conflict Risk	The description is anchored to a specific product (LaunchDarkly) and a specific activity (creating/configuring flags), with triggers like "wrap code in a flag" that would not fire for unrelated skills. It is clearly distinguishable from a sibling flag-targeting skill because the stated scope is creation, not rollout/targeting; not level 4 since no meaningful overlap risk is evident.	5 / 5
	Total	18 / 20 Passed

Validation

93%

Warnings & errors only

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 15 / 16 Passed

Validation for skill structure

Criteria	Description	Result
relative_links	Relative link issues: 4 suspicious	Warning

	Total	15 / 16 Passed

Repository: launchdarkly/ai-tooling
Path: skills/feature-flags/launchdarkly-flag-create/SKILL.md
Commit: ef54971

Reviewed: 6 days ago

Table of Contents

Discovery Implementation Validation

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.