# API Latency Breakdown

API latency breakdown monitor showing network time vs processing time split, p95 latency tracking, and performance bottleneck detection for Claude Code sessions.

---


## Metadata

**Title:** API Latency Breakdown
**Category:** statuslines
**Author:** JSONbored
**Added:** October 2025
**Tags:** api-latency, performance, monitoring, bottleneck, network
**URL:** https://claudepro.directory/statuslines/api-latency-breakdown

## Overview

API latency breakdown monitor showing network time vs processing time split, p95 latency tracking, and performance bottleneck detection for Claude Code sessions.

## Content

#!/usr/bin/env bash
API LATENCY BREAKDOWN MONITOR FOR CLAUDE CODE
SHOWS NETWORK/WAITING TIME VS ACTUAL API PROCESSING TIME
READ JSON FROM STDIN
read -r input
EXTRACT VALUES
totaldurationms=$(echo "$input" | jq -r '.cost.totaldurationms // 0')
apidurationms=$(echo "$input" | jq -r '.cost.totalapiduration_ms // 0')
CALCULATE NETWORK/WAITING TIME (TOTAL - API)
networktime=$((totaldurationms - apiduration_ms))
AVOID NEGATIVE VALUES
if [ $network_time -lt 0 ]; then
 network_time=0
fi
CONVERT TO SECONDS FOR DISPLAY
apiseconds=$(echo "scale=2; $apiduration_ms / " | bc)
networkseconds=$(echo "scale=2; $networktime / " | bc)
totalseconds=$(echo "scale=2; $totalduration_ms / " | bc)
CALCULATE PERCENTAGE SPLIT
if [ $totaldurationms -gt 0 ]; then
 apipercentage=$(( (apidurationms * ) / totalduration_ms ))
 networkpercentage=$(( - apipercentage ))
else
 api_percentage=0
 network_percentage=0
fi
PERFORMANCE ASSESSMENT (NETWORK TIME SHOULD BE MINIMAL)
if [ $network_time -lt ]; then # 5s network time)
? Visual ratio bar (blue = API processing, orange = network overhead)
? Helps identify network issues vs API performance problems
? Lightweight bash with bc for floating-point calculations
CONFIGURATION
Format: bash
Refresh Interval: 1000ms
Position: left
USE CASES
? Performance debugging for slow Claude Code responses
? Identifying network bottlenecks vs API processing delays
? Optimizing API call efficiency in distributed teams
? Troubleshooting VPN/proxy latency issues
? Monitoring API performance degradation over time
? Production environment SLA monitoring
TROUBLESHOOTING
1) Network time showing negative or zero despite slow responses
 Solution: Verify both cost.totaldurationms and cost.totalapidurationms fields exist: echo '$input' | jq .cost. Negative protection caps networktime at 0. If both values missing, script shows 0s - check Claude Code version supports these fields.
2) API percentage always showing % with no network time
 Solution: This indicates totaldurationms equals totalapidurationms (no measurable network overhead). Verify calculations: networktime = total - API. If consistently 0, either network is extremely fast or fields are identical in JSON. Check actual JSON values.
3) Performance status showing BOTTLENECK incorrectly
 Solution: Thresholds: 5s red (BOTTLENECK). Adjust thresholds in script if your network baseline differs. VPN users may see higher normal latency. Modify: network_time -lt to higher value for VPN environments.
4) Ratio bar not displaying or showing as empty
 Solution: Bar uses ANSI background colors (\[48;5;Xm). Ensure terminal supports -color mode: tput colors (should return ). If bars show as spaces only, verify ANSI escape codes working: echo -e '\[48;5;75m BLUE \[0m'.
5) bc: command not found when calculating percentages
 Solution: Install bc: brew install bc (macOS), apt install bc (Linux). Alternative: use integer math only: apipercentage=$((apidurationms * / totalduration_ms)) - removes decimal precision but works without bc.
TECHNICAL DETAILS
Documentation: https://docs.claude.com/en/docs/claude-code/statusline
PREVIEW
✓ FAST | API: 2.34s (85%) | Network: 0.41s (15%) | █████████████████░░░


---

Source: Claude Pro Directory
Website: https://claudepro.directory
URL: https://claudepro.directory/statuslines/api-latency-breakdown

This content is optimized for Large Language Models (LLMs).
For full formatting and interactive features, visit the website.