Name: configs-create
Rating: 64 (1 reviews)
Author: launchdarkly

configs-create

Create and configure configs in LaunchDarkly. Helps you choose between agent vs completion mode, create the config, add variations with models and prompts, and verify the setup.

Quality

80%

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

Fix and improve this skill with Tessl

tessl review fix ./skills/agentcontrol/configs-create/SKILL.md

Quality

Content

88%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.

The body is an exemplar of actionable, domain-specific guidance: every step is executable, validation checkpoints and failure modes are explicit, and the knowledge conveyed (fallthrough semantics, modelName quirk, broken semantic-patch instruction) is genuinely non-obvious. Its only real weaknesses are a thrice-repeated warning that could be consolidated and some reference-worthy material inlined in a monolithic file.

Suggestions

Consolidate the fallthrough/'enabled=False' warning into Step 5 and reference it briefly from the intro blockquote and the 'What NOT to Do' list to cut redundant tokens.

Move the direct curl/ldcli targeting shortcut and the SDK reference link list into a references/ file (e.g., TARGETING-SHORTCUT.md) and link to it from Step 5, keeping SKILL.md a lean overview.

Trim the 'More resources' section to one or two canonical links, since the six SDK-specific links add token cost for a rarely needed branch.

Dimension	Reasoning	Score
Conciseness	Nearly every section conveys non-obvious operational knowledge Claude cannot infer (fallthrough defaults to a disabled variation, 'modelName' is required at create time, 'turnTargetingOn' does not work for configs), but the fallthrough warning is repeated nearly verbatim three times (intro blockquote, Step 5, 'What NOT to Do') and could be consolidated. Not 5 due to that redundancy; well above level 3 since no section explains concepts Claude already knows.	4 / 5
Actionability	Fully executable, copy-paste-ready guidance: complete JSON payloads for both agent and completion modes, exact curl and ldcli commands with required headers and the semantic-patch body, the config URL pattern, and exact field names ('modelConfigKey', 'modelName'). Covers the common cases for both the one-step and two-step creation paths.	5 / 5
Workflow Clarity	Steps 1-5 are clearly sequenced with explicit validation checkpoints (Step 4's four-point verification checklist, the 'NO MODEL' check, the post-targeting 'enabled is True' assertion), an edge-case table, and feedback loops for error recovery (fix jq shape by inspecting first, use update-ai-config-variation for missing models). The servability step is explicitly called out as mandatory before completion.	5 / 5
Progressive Disclosure	The single SKILL.md file is well organized with clear section headers and one-level-deep external doc links, and no bundle files exist to offload content. Not 5 because the skill exceeds 50 lines and inlines material that could live in reference files (the direct curl/CLI targeting shortcut, the SDK resource link list), which keeps it slightly monolithic.	4 / 5
	Total	18 / 20 Passed

Description

61%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.

The description clearly conveys what the skill does with a good list of concrete actions and relevant LaunchDarkly trigger terms, but it lacks any 'use when' guidance and slips into second-person voice ('Helps you choose'). Adding an explicit trigger clause and third-person phrasing would lift it substantially.

Suggestions

Add an explicit 'Use when...' clause (e.g., 'Use when the user wants to create a new LaunchDarkly AI config, set up agent or completion mode, or add the first variation with a model and prompt') to satisfy the completeness cap.

Rewrite in third person: replace 'Helps you choose between agent vs completion mode' with 'Guides the choice between agent vs completion mode' to remove the second-person penalty.

Include natural synonyms users say, such as 'AI config' and 'model config', to broaden trigger-term coverage and reduce confusion with generic 'config' skills.

Dimension	Reasoning	Score
Specificity	Names the domain and several concrete actions ('create the config, add variations with models and prompts, and verify the setup') which fits the level-4 anchor, but the second-person phrasing 'Helps you choose' violates the third-person voice rule and incurs the -1 specificity penalty.	3 / 5
Completeness	The 'what' is clearly stated (create and configure configs, choose mode, add variations, verify), but there is no 'Use when...' clause or equivalent explicit trigger guidance, which the judging guidelines cap at 3.	3 / 5
Trigger Term Quality	Includes good natural keywords users would say ('LaunchDarkly', 'configs', 'agent vs completion mode', 'variations', 'models', 'prompts'), but misses common synonyms like 'AI configs' or 'AI flags'. Not level 5 because synonym and variation coverage is incomplete; not level 3 because the core terms users would naturally use when requesting this skill are present.	4 / 5
Distinctiveness Conflict Risk	The LaunchDarkly AI-config niche and the 'create' verb make it mostly distinct, but 'and configure configs' creates minor overlap risk with the closely related configs-update and configs-targeting sibling skills. Not 5 because of that overlap; not 3 because the creation framing and mode-choice terms are distinguishing.	4 / 5
	Total	14 / 20 Passed

Validation

100%

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 — 16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository: launchdarkly/ai-tooling
Path: skills/agentcontrol/configs-create/SKILL.md
Commit: ef54971

Reviewed: 6 hours 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.