CtrlK
BlogDocsLog inGet started
Tessl Logo

migrate

Migrate an application with hardcoded LLM prompts to a full LaunchDarkly AgentControl implementation in five stages: audit the code, wrap the call, move the tools, add tracking, attach evaluators. Use when the user wants to externalize model/prompt configuration, move from direct provider calls (OpenAI, Anthropic, Bedrock, Gemini, Strands) to a managed config, or stage a full hardcoded-to-LaunchDarkly migration.

69

Quality

85%

Does it follow best practices?

Impact

No eval scenarios have been run

SecuritybySnyk

Advisory

Suggest reviewing before use

SKILL.md
Quality
Evals
Security

Quality

Content

70%

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

This is a highly actionable and well-structured migration skill with excellent workflow clarity, explicit validation checkpoints, and proper progressive disclosure to reference files. Its primary weakness is extreme verbosity — the same warnings and rules are repeated across multiple sections (callout box, workflow steps, What NOT to Do, Edge Cases), and many explanations could be dramatically condensed given Claude's existing knowledge. The coverage matrix and edge case table, while useful, could be moved to reference files to reduce the main skill's token footprint.

Suggestions

Move the coverage matrix table and the large edge cases table to reference files (e.g., coverage-matrix.md and edge-cases.md) and link to them from the main skill — this alone would cut ~100 lines.

Consolidate the 'What NOT to Do' section by removing items that are already stated in the workflow steps (e.g., tracker lifetime rules are explained in Stage 4 Step 1 and repeated nearly verbatim in What NOT to Do); keep only a brief cross-reference.

Remove explanatory commentary that Claude doesn't need (e.g., 'PDF-style' explanations of what offline mode does, why AttributeError occurs, what at-most-once semantics mean) and trust Claude to understand the API contracts from the code examples alone.

Trim inline code comments that restate the surrounding prose (e.g., '# .format() is removed at the call site — the SDK interpolates via variables' when the prose already explains this).

DimensionReasoningScore

Conciseness

This skill is extremely verbose at ~700+ lines. It over-explains concepts Claude already knows (e.g., what offline mode is, how Python imports work, what AttributeError means), repeats the same warnings multiple times across sections (tracker lifetime rules appear in Stage 4, What NOT to Do, and the callout box), and includes extensive coverage matrices and edge case tables that could be in reference files. The 'What NOT to Do' section alone is massive and largely restates guidance already given in the workflow.

1 / 3

Actionability

The skill provides fully executable, copy-paste-ready code examples for both Python and Node.js across every stage. Specific commands for package installation, concrete before/after code transformations, exact API calls with correct parameter names, and structured output templates are all present. The guidance is highly specific and leaves little ambiguity about what to do.

3 / 3

Workflow Clarity

The five-stage workflow is clearly sequenced with explicit validation checkpoints at each stage (sub-step 9 in Stage 2, sub-step 4 in Stage 3, sub-step 5 in Stage 4, sub-step 5 in Stage 5). The Stage 1 checkpoint with four explicit confirmation forms is particularly well-designed. Feedback loops are present (e.g., 'If errors: fix and re-validate'). The hand-off model between sibling skills is clearly documented with explicit 'do not auto-invoke' guards.

3 / 3

Progressive Disclosure

The skill has a clear overview structure with well-signaled one-level-deep references to supporting files (phase-1-analysis-checklist.md, before-after-examples.md, sdk-ai-tracker-patterns.md, agent-mode-frameworks.md, fallback-defaults-pattern.md, agent-graph-reference.md). The References section at the bottom provides a clean navigation index. Content is appropriately split between the main workflow and detailed reference materials.

3 / 3

Total

10

/

12

Passed

Description

100%

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

This is an excellent skill description that clearly articulates a specific migration workflow with concrete stages, includes a well-formed 'Use when' clause with multiple trigger scenarios, and names specific technologies that serve as strong natural keywords. The description is concise yet comprehensive, making it easy for Claude to distinguish this skill from others in a large skill set.

DimensionReasoningScore

Specificity

Lists five specific concrete stages (audit the code, wrap the call, move the tools, add tracking, attach evaluators) and names the overall action (migrate an application with hardcoded LLM prompts to LaunchDarkly AgentControl). Very concrete and actionable.

3 / 3

Completeness

Clearly answers both 'what' (migrate hardcoded LLM prompts to LaunchDarkly AgentControl in five stages) and 'when' (explicit 'Use when' clause covering externalize config, move from direct provider calls, or stage a full migration). Both are well-articulated.

3 / 3

Trigger Term Quality

Includes strong natural trigger terms: 'externalize model/prompt configuration', 'direct provider calls', specific provider names (OpenAI, Anthropic, Bedrock, Gemini, Strands), 'hardcoded-to-LaunchDarkly migration', 'managed config'. These cover multiple natural ways a user might describe this need.

3 / 3

Distinctiveness Conflict Risk

Highly distinctive — targets a very specific niche (LaunchDarkly AgentControl migration for LLM prompts) with named providers and a clear five-stage process. Unlikely to conflict with other skills given the specificity of the domain and tooling.

3 / 3

Total

12

/

12

Passed

Validation

90%

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

Validation10 / 11 Passed

Validation for skill structure

CriteriaDescriptionResult

skill_md_line_count

SKILL.md is long (573 lines); consider splitting into references/ and linking

Warning

Total

10

/

11

Passed

Repository
launchdarkly/ai-tooling
Reviewed

Table of Contents

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.