langfuse-observation-view

📁 neuradex/langfuse-skills 📅 5 days ago
4
总安装量
2
周安装量
#52833
全站排名
安装命令
npx skills add https://github.com/neuradex/langfuse-skills --skill langfuse-observation-view

Agent 安装分布

claude-code 2
amp 1
opencode 1
cursor 1
kimi-cli 1
codex 1

Skill 文档

Langfuse Observation View

Display detailed information for a specific Observation (Generation/Span).

Setup

Set the following environment variables before use:

Variable Required Description
LANGFUSE_PUBLIC_KEY Yes Langfuse public key
LANGFUSE_SECRET_KEY Yes Langfuse secret key
LANGFUSE_HOST or LANGFUSE_BASE_URL No Langfuse host URL (default: https://us.cloud.langfuse.com)

When to Use

  • Checking specific LLM call details
  • Viewing input/output
  • Checking costs and token usage
  • Viewing prompts used

Commands

1. Get Observation Details

npx tsx scripts/langfuse-observation-view.ts <observationId>

2. Show Full Input/Output

npx tsx scripts/langfuse-observation-view.ts <observationId> --full

3. JSON Output

npx tsx scripts/langfuse-observation-view.ts <observationId> --json

Output Example

Langfuse Observation Detail
===========================

ID:          obs_abc123
Trace ID:    trace_xyz789
Type:        GENERATION
Name:        chat
Start Time:  2026-01-21 10:30:45
Duration:    2.35s
Level:       DEFAULT

--- Model Info ---
Model:       qwen/qwen3-32b
Parameters:  {"temperature":0.7}

--- Usage & Cost ---
Input Tokens:   1500
Output Tokens:  350
Total Tokens:   1850
Input Cost:     $0.000435
Output Cost:    $0.000207
Total Cost:     $0.000642

--- Prompt ---
Prompt:      librarian-system (v3)
Prompt ID:   prompt_def456

--- Input ---
[{"role":"system","content":"You are..."}]

--- Output ---
Here are the search results...

Fields

Field Description
Type GENERATION (LLM call), SPAN (processing span), EVENT (event)
Duration Processing time
Model Model used
Input/Output Tokens Token usage
Calculated Cost Computed cost
Prompt Langfuse prompt used