CtrlK
BlogDocsLog inGet started
Tessl Logo

llmobs-integration

Use when adding, debugging, or modifying LLMObs plugins for an LLM library in dd-trace-js. Triggers: "add LLMObs support", "instrument chat completions / streaming / embeddings / agent runs / orchestration / tool calls / retrieval", "LLMObsPlugin", "getLLMObsSpanRegisterOptions", "setLLMObsTags", "SPAN_KINDS", "span kind", any provider tag ("openai" / "anthropic" / "genai" / "google" / "langchain" / "langgraph" / "ai" llmobs), "VCR cassettes".

68

Quality

84%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

SecuritybySnyk

Passed

No findings from the security scan

SKILL.md
Quality
Evals
Security

Quality

Content

68%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 well-organized, concise, and actionable with good progressive disclosure into existing reference files. Its main weakness is workflow_clarity: the implementation sequence lacks an explicit validation step for a batch instrumentation task.

Suggestions

Add an explicit validation step to the Implementation Steps workflow, e.g. 'Run the VCR cassettes / LLMObs tests and confirm span events emit with the expected kind, input, output, and metadata' before declaring the plugin done.

Link the orphaned references/reference-implementations.md from the body (e.g., in the Implementation Steps or Plugin Architecture section) so all bundle files are discoverable.

Trim the justificatory asides (e.g., the 'LLM libraries iterate fast' preamble and the 'not constants in the codebase' caveat) to tighten conciseness toward the lean top anchor.

DimensionReasoningScore

Conciseness

Mostly efficient and assumes Claude's competence, conveying domain-specific knowledge (package shapes, span kinds, message extraction) without explaining basics; a few justificatory asides ('LLM libraries iterate fast...', 'not constants in the codebase') could be trimmed, keeping it just below the lean top anchor.

4 / 5

Actionability

Provides concrete guidance — specific file paths, method signatures, field lists, and a numbered 5-step implementation sequence — that is actionable even without inline runnable code, which is acceptable for an instruction-heavy skill, though the absence of any copy-paste example is a minor gap.

4 / 5

Workflow Clarity

The Implementation Steps section gives a clear sequenced process, but it lacks an explicit validation/verification checkpoint (e.g., running VCR cassettes or confirming spans emit), and the missing feedback loop for a batch instrumentation operation caps this at 3 per the rubric guidance.

3 / 5

Progressive Disclosure

The body is well-structured with clear sections and well-signaled one-level-deep markdown links to real bundle files (plugin-architecture.md, category-detection.md, message-extraction.md), but references/reference-implementations.md exists in the bundle yet is never linked from the body, a minor navigation gap.

4 / 5

Total

15

/

20

Passed

Description

100%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 is specific, complete, and rich in natural trigger terms with a clear 'Use when' clause. It carves out a distinct niche with minimal conflict risk.

DimensionReasoningScore

Specificity

Names the domain and lists multiple concrete actions ('adding, debugging, or modifying LLMObs plugins') plus comprehensive instrumented operations (chat completions, streaming, embeddings, agent runs, orchestration, tool calls, retrieval).

5 / 5

Completeness

Clearly answers both 'what' (create/debug/modify LLMObs plugins instrumenting LLM library operations) and 'when' with concrete trigger phrases, matching the top anchor.

5 / 5

Trigger Term Quality

Includes an explicit 'Use when...' clause followed by a 'Triggers:' list with natural phrases, provider-name synonyms (openai, anthropic, genai, google, langchain, langgraph), and technical identifiers, giving comprehensive keyword coverage.

5 / 5

Distinctiveness Conflict Risk

Occupies a clear niche (LLMObs plugins for dd-trace-js) with distinctive technical triggers (LLMObsPlugin, getLLMObsSpanRegisterOptions, SPAN_KINDS) that minimize conflict with other skills.

5 / 5

Total

20

/

20

Passed

Validation

93%

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

Validation15 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

relative_links

Relative link issues: 2 suspicious

Warning

Total

15

/

16

Passed

Repository
DataDog/dd-trace-js
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.