Back to Skills

config

majiayu000
Updated Yesterday
1 views
58
9
58
View on GitHub
Developmentaiapi

About

The config skill manages Claude Code's configuration settings, including API keys and optional features like Braintrust tracing. It provides commands to enable/disable Braintrust integration and check configuration status. Use this skill when you need to modify tracing settings or verify your current Claude Code setup.

Quick Install

Claude Code

Recommended
Plugin CommandRecommended
/plugin add https://github.com/majiayu000/claude-skill-registry
Git CloneAlternative
git clone https://github.com/majiayu000/claude-skill-registry.git ~/.claude/skills/config

Copy and paste this command in Claude Code to install this skill

Documentation

Config - Configuration Management

Manage Clorch optional features and API keys.

Usage

/config                    # Show configuration menu
/config braintrust         # Braintrust settings
/config braintrust on      # Enable Braintrust tracing
/config braintrust off     # Disable Braintrust tracing
/config braintrust status  # Check Braintrust status

Braintrust Configuration

Check Status

# Check current Braintrust configuration
grep -E "BRAINTRUST|TRACE_TO" ~/.claude/.env 2>/dev/null || echo "Not configured"

Enable Braintrust

If API key exists:

# Update .env to enable
sed -i '' 's/TRACE_TO_BRAINTRUST=false/TRACE_TO_BRAINTRUST=true/' ~/.claude/.env 2>/dev/null || \
  echo "TRACE_TO_BRAINTRUST=true" >> ~/.claude/.env
echo "Braintrust tracing enabled. Restart Claude Code to apply."

If no API key, prompt user:

question: "Braintrust API key not found. How would you like to proceed?"
header: "Setup"
options:
  - label: "Enter API key now"
    description: "I have a Braintrust API key ready"
  - label: "Get an API key first"
    description: "Open braintrust.dev to sign up (free tier available)"
  - label: "Skip for now"
    description: "I'll set it up later"

Disable Braintrust

# Update .env to disable
sed -i '' 's/TRACE_TO_BRAINTRUST=true/TRACE_TO_BRAINTRUST=false/' ~/.claude/.env
echo "Braintrust tracing disabled. Restart Claude Code to apply."

Interactive Menu

When user runs just /config:

question: "What would you like to configure?"
header: "Config"
options:
  - label: "Braintrust (observability)"
    description: "Token tracking, cost analysis, session debugging"
  - label: "API Keys"
    description: "Perplexity, Nia, and other service keys"
  - label: "Show current config"
    description: "Display all configuration values"

Braintrust Submenu

question: "Braintrust configuration:"
header: "Braintrust"
options:
  - label: "Enable tracing"
    description: "Turn on session tracking"
  - label: "Disable tracing"
    description: "Turn off (keeps API key)"
  - label: "Update API key"
    description: "Change or add API key"
  - label: "View status"
    description: "Show current Braintrust config"

Benefits Display

When explaining Braintrust:

## What is Braintrust?

Braintrust is an LLM observability platform.

### Benefits:
- Token tracking: See exactly how many tokens each task uses
- Cost analysis: Track API costs across sessions
- Performance debugging: Identify slow agents
- Session replay: Review what happened in past sessions

### Data sent:
- Session metadata (timestamps, agent IDs)
- Token usage per request
- Tool calls and durations
- Conversation context (first 100 chars)

### Privacy:
- Your Braintrust account, your data
- API key stays local in ~/.claude/.env
- Disable anytime: /config braintrust off

Get a free API key: https://braintrust.dev

Implementation

Parse Command

args = user_input.split()
# /config -> show menu
# /config braintrust -> braintrust menu
# /config braintrust on -> enable
# /config braintrust off -> disable
# /config braintrust status -> show status

Toggle Function

toggle_braintrust() {
    local action=$1
    local env_file="$HOME/.claude/.env"

    case $action in
        on)
            if grep -q "BRAINTRUST_API_KEY=" "$env_file" 2>/dev/null; then
                sed -i '' 's/TRACE_TO_BRAINTRUST=false/TRACE_TO_BRAINTRUST=true/' "$env_file"
                echo "Enabled. Restart Claude Code to apply."
            else
                echo "No API key found. Run /config braintrust to set up."
            fi
            ;;
        off)
            sed -i '' 's/TRACE_TO_BRAINTRUST=true/TRACE_TO_BRAINTRUST=false/' "$env_file"
            echo "Disabled. Restart Claude Code to apply."
            ;;
        status)
            if grep -q "TRACE_TO_BRAINTRUST=true" "$env_file" 2>/dev/null; then
                echo "Braintrust: ENABLED"
            else
                echo "Braintrust: DISABLED"
            fi
            grep "BRAINTRUST_API_KEY=" "$env_file" 2>/dev/null | sed 's/=.*/=***/' || echo "API Key: Not set"
            ;;
    esac
}

Quick Reference

CommandAction
/configInteractive config menu
/config braintrustBraintrust setup wizard
/config braintrust onEnable tracing
/config braintrust offDisable tracing
/config braintrust statusShow status

GitHub Repository

majiayu000/claude-skill-registry
Path: skills/config

Related Skills

creating-opencode-plugins

Meta

This skill provides the structure and API specifications for creating OpenCode plugins that hook into 25+ event types like commands, files, and LSP operations. It offers implementation patterns for JavaScript/TypeScript modules that intercept and extend the AI assistant's lifecycle. Use it when you need to build event-driven plugins for monitoring, custom handling, or extending OpenCode's capabilities.

View skill

sglang

Meta

SGLang is a high-performance LLM serving framework that specializes in fast, structured generation for JSON, regex, and agentic workflows using its RadixAttention prefix caching. It delivers significantly faster inference, especially for tasks with repeated prefixes, making it ideal for complex, structured outputs and multi-turn conversations. Choose SGLang over alternatives like vLLM when you need constrained decoding or are building applications with extensive prefix sharing.

View skill

evaluating-llms-harness

Testing

This Claude Skill runs the lm-evaluation-harness to benchmark LLMs across 60+ standardized academic tasks like MMLU and GSM8K. It's designed for developers to compare model quality, track training progress, or report academic results. The tool supports various backends including HuggingFace and vLLM models.

View skill

langchain

Meta

LangChain is a framework for building LLM applications using agents, chains, and RAG pipelines. It supports multiple LLM providers, offers 500+ integrations, and includes features like tool calling and memory management. Use it for rapid prototyping and deploying production systems like chatbots, autonomous agents, and question-answering services.

View skill