Name: langfuse
Rating: 54.400000000000006 (1 reviews)
Author: administrakt0r

langfuse

You are an expert in LLM observability and evaluation. You think in terms of traces, spans, and metrics. You know that LLM applications need monitoring just like traditional software - but with different dimensions (cost, quality, latency).

Quality

61%

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 ./plugins/AI-Agents-Safe-Coding-Skills-claude/skills/langfuse/SKILL.md

The canonical home for this skill is langfuse in administrakt0r/AI-Agents-Safe-Coding-Skills

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.

The body is a well-structured, actionable reference with three solid executable integration patterns and useful anti-pattern guidance. Its main weaknesses are minor: a few undefined variables in code, restated persona prose, and a vague closing 'When to Use' line.

Suggestions

Fix the small code gaps so examples are fully runnable: add the missing `import openai` in the basic example and define or stub `tools`/`prompt` in the LangChain agent snippet.

Replace the filler closing line ('This skill is applicable to execute the workflow or actions described in the overview') with a concrete 'When to Use' statement listing real trigger scenarios.

Drop the redundant 'You are an expert...' opening paragraph since it duplicates the frontmatter description and adds tokens without new information.

Dimension	Reasoning	Score
Conciseness	The body is mostly efficient: code examples carry the load with minimal prose, and anti-patterns use tight 'Why bad'/'Instead' framing. Minor trimming is possible — the 'You are an expert...' intro restates the frontmatter and the closing 'This skill is applicable to execute the workflow or actions described in the overview' is vague filler — so it sits at 4 rather than 5.	4 / 5
Actionability	Three complete, mostly copy-paste-ready Python examples (Langfuse client, `langfuse.openai` drop-in, LangChain `CallbackHandler`) cover the common integrations, but minor gaps keep it at 4: the first example omits `import openai` and the LangChain agent snippet references undefined `tools`/`prompt` variables.	4 / 5
Workflow Clarity	The basic-tracing example lays out a clear linear sequence (init → trace → generation → end → score → flush) with a noted flush checkpoint for serverless, and each pattern carries a 'When to use' line. It is not 5 because there are no explicit validate/fix/retry feedback loops, though the non-destructive nature means those are not strictly required.	4 / 5
Progressive Disclosure	Content is organized into clear, well-labeled sections (Capabilities, Requirements, Patterns, Anti-Patterns, Limitations, Related Skills) with no bundle files present and no broken or deeply nested references. It is not 5 because the three full integration examples are inlined where a reference file split could ease navigation, but the structure is sound and self-contained.	4 / 5
	Total	16 / 20 Passed

Description

48%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 establishes a clear LLM-observability niche but is written as a second-person persona with no 'Use when' trigger guidance, which caps completeness and lowers specificity per the rubric. It is functional but would benefit from a third-person, action-oriented rewrite with explicit trigger terms.

Suggestions

Rewrite in third person with concrete verbs (e.g. 'Traces LLM calls, scores outputs, and tracks cost/quality/latency') instead of the 'You are an expert... You think...' persona framing.

Add an explicit 'Use when...' clause naming natural user triggers (e.g. 'Use when monitoring LLM applications, tracing LLM calls, or evaluating LLM output quality').

Include common synonyms and the product name as triggers (e.g. 'LLM monitoring', 'tracing LLM calls', 'Langfuse') to improve natural-keyword coverage.

Dimension	Reasoning	Score
Specificity	The description names the domain ('LLM observability and evaluation') and concrete concepts ('traces, spans, and metrics', 'cost, quality, latency') but describes a persona ('You think in terms of', 'You know that') rather than concrete actions; the second-person voice ('You think', 'You know') triggers the mandated -1 specificity penalty, bringing an already action-light base of 3 down to 2.	2 / 5
Completeness	A reasonably clear 'what' is present (LLM observability/evaluation across cost, quality, latency) but there is no 'when'/'Use when' clause, so per the rubric a missing explicit trigger caps completeness at 3.	3 / 5
Trigger Term Quality	Relevant keywords appear ('monitoring', 'traces', 'evaluation', 'observability') but the phrasing leans technical and omits natural user variations like 'trace my LLM calls' or 'monitor my LLM app', and there is no explicit trigger phrase.	3 / 5
Distinctiveness Conflict Risk	The niche (LLM observability, tracing/spans/metrics) is fairly distinct from most skills with only minor overlap risk against general monitoring or evaluation skills; it is not a 5 because no explicit trigger phrase sharpens the boundary.	4 / 5
	Total	12 / 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
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning

	Total	15 / 16 Passed

Repository: administrakt0r/AI-Agents-Safe-Coding-Skills
Path: plugins/AI-Agents-Safe-Coding-Skills-claude/skills/langfuse/SKILL.md
Commit: 57c135b

Reviewed: 4 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.