charlesjones-dev/claude-code-plugins-dev/statusline-edit

Status Line Edit

Edit your existing Claude Code status line configuration.

Instructions

CRITICAL: This command MUST NOT accept any arguments. If the user provided any text after this command, COMPLETELY IGNORE it.

Phase 1: Detect Existing Configuration

1. Detect the operating system using Bash: uname -s

• "Darwin" = macOS, "Linux" = Linux, otherwise assume Windows

1. Check if the status line script exists:

• Mac/Linux: ~/.claude/statusline.sh
• Windows: C:/Users/USERNAME/.claude/statusline.ps1

1. If the script does NOT exist, display:

``` No status line script found at ~/.claude/statusline.sh

Run /statusline-wizard to set up your status line first. ``` Then STOP - do not continue.

Phase 2: Read Current Configuration

Read the existing script file and parse the current SHOW_* variable values:

For Bash scripts, look for lines like:

SHOW_MODEL=true
SHOW_EFFORT=true
SHOW_SANDBOX=false
SHOW_TOKEN_COUNT=true
SHOW_PROGRESS_BAR=true
SHOW_DIRECTORY=true
SHOW_GIT_BRANCH=true
SHOW_COST=false
SHOW_DURATION=true
SHOW_TIME=true
SHOW_VERSION=true
SHOW_RATE_LIMITS=true
SHOW_SPEND_BUDGET=false
SPEND_LIMIT_USD=0
SPEND_RESET_DAY=1
SPEND_RESET_TIME=00:00

For PowerShell scripts, look for lines like:

$SHOW_MODEL = $true
$SHOW_EFFORT = $true
$SHOW_TOKEN_COUNT = $true
$SHOW_PROGRESS_BAR = $true
$SHOW_DIRECTORY = $true
$SHOW_GIT_BRANCH = $true
$SHOW_COST = $false
$SHOW_DURATION = $true
$SHOW_TIME = $true
$SHOW_VERSION = $true
$SHOW_RATE_LIMITS = $true
$SHOW_SPEND_BUDGET = $false
$SPEND_LIMIT_USD = 0
$SPEND_RESET_DAY = 1
$SPEND_RESET_TIME = '00:00'

Store the current values to use as defaults in the wizard.

Note: Scripts generated by older plugin versions may lack SHOW_EFFORT (pre-1.3.0), SHOW_SPEND_BUDGET / SPEND_LIMIT_USD / SPEND_RESET_DAY / SPEND_RESET_TIME (pre-1.4.0), or SHOW_SANDBOX (pre-1.5.0). If a variable is not present in the script, OMIT the matching option from the wizard (toggling it would have no effect) and, after the update completes, mention that re-running /statusline-wizard will regenerate the script with support for the missing segment(s). The exception is SHOW_SANDBOX in a PowerShell script: it's never there, because the sandbox doesn't run on native Windows, so omit the option without suggesting a re-run.

Scripts generated before 1.5.2 have older rate-limit code. If the script has SHOW_RATE_LIMITS but no rate_limit_pct function (Bash) or Get-RateLimitPercent function (PowerShell), keep the option, and after the update mention that re-running /statusline-wizard updates the rate-limit code. The newer code leaves out a rate-limit window that is empty or malformed, instead of showing 5h:0% or printing errors, which matters on accounts without a 5-hour limit.

Phase 3: Configuration Wizard with Pre-selected Values

Use AskUserQuestion with these grouped questions. Pre-select options based on current values from Phase 2.

Question 1 - Context Display (multiSelect: true): Options (pre-select based on current config):

• Model name - select if SHOW_MODEL=true
• Effort level - select if SHOW_EFFORT=true
• Token count (50k/100k) - select if SHOW_TOKEN_COUNT=true
• Progress bar - select if SHOW_PROGRESS_BAR=true

Question 2 - Project Display (multiSelect: true): Options (pre-select based on current config):

• Current directory - select if SHOW_DIRECTORY=true
• Git branch - select if SHOW_GIT_BRANCH=true
• Sandbox indicator - select if SHOW_SANDBOX=true. Description: "Shows 'Sandbox' in orange after the effort level while Claude Code's Bash sandbox is on for this session. Hidden when it's off."

Question 3 - Session Display (multiSelect: true): Options (pre-select based on current config):

• Session duration - select if SHOW_DURATION=true
• Current time - select if SHOW_TIME=true
• Claude Code version - select if SHOW_VERSION=true
• Session cost - select if SHOW_COST=true

Question 4 - Usage Limits (multiSelect: true): Options (pre-select based on current config):

• Rate limit usage (5h/7d) - select if SHOW_RATE_LIMITS=true. Description: "Only shown on plans with rolling rate-limit windows (Pro, Max and Ultra). Each window shows only when your plan has it, so an account without a 5-hour limit shows 7d alone. Enterprise and API seats billed against a monthly spend cap have no rate limits, so the segment stays hidden. Pick Monthly spend budget instead."
• Monthly spend budget (mo:$47.20/$2000 2.4%) - select if SHOW_SPEND_BUDGET=true. Description: "Month-to-date spend across all Claude Code sessions on this machine, as estimated locally by Claude Code."

Questions 5 and 6 - Spend cap and reset (only ask if "Monthly spend budget" is selected; send both in one AskUserQuestion call, single select each):

Question 5 - Monthly spend cap (USD): offer "No cap - show bare total", "$500", "$1000", "$2000", pre-selecting the option that matches the current SPEND_LIMIT_USD (or describe the current custom value in the question text). The user can pick "Other" and type any amount; write the numeric value (digits and optional decimal point only) into SPEND_LIMIT_USD. 0 means no cap.

Question 6 - When does the cap reset?: offer "1st of the month at 00:00 UTC" (Anthropic console default), "1st of the month at midnight local time", and "Keep current: day N at HH:MM UTC" showing the values read in Phase 2. The user can pick "Other" and paste what the console shows (e.g., "Resets Wed, Sep 30, 8:00 PM EDT"); convert it to UTC and write the day of month (1-28) to SPEND_RESET_DAY and HH:MM (24-hour) to SPEND_RESET_TIME. Tell the user the UTC values you wrote.

Phase 4: Update Script

Update ONLY the SHOW_* variables (and SPEND_LIMIT_USD, SPEND_RESET_DAY, SPEND_RESET_TIME) at the top of the existing script file based on wizard selections.

For Bash: Use Edit tool to replace the configuration block:

SHOW_MODEL=true           # Show model name (e.g., "Claude Opus 4.8")
SHOW_EFFORT=true          # Show reasoning effort level (e.g., "high")
SHOW_SANDBOX=false        # Show "Sandbox" when Claude Code's Bash sandbox is enabled for this session
SHOW_TOKEN_COUNT=true     # Show token usage count (e.g., "50k/100k")
...

For PowerShell: Use Edit tool to replace the configuration block:

$SHOW_MODEL = $true           # Show model name (e.g., "Claude Opus 4.8")
$SHOW_EFFORT = $true          # Show reasoning effort level (e.g., "high")
$SHOW_TOKEN_COUNT = $true     # Show token usage count (e.g., "50k/100k")
...

IMPORTANT: Do NOT regenerate the entire script. Only update the configuration variables section.

Success Message

After successful update, display:

Status line updated!
 
Check out your refreshed status line below!
 
Current configuration:
- Model name: [enabled/disabled]
- Effort level: [enabled/disabled]
- Sandbox indicator: [enabled/disabled] (omit this line on Windows)
- Token count: [enabled/disabled]
- Progress bar: [enabled/disabled]
- Directory: [enabled/disabled]
- Git branch: [enabled/disabled]
- Session cost: [enabled/disabled]
- Duration: [enabled/disabled]
- Time: [enabled/disabled]
- Version: [enabled/disabled]
- Rate limits: [enabled/disabled]
- Monthly spend budget: [enabled/disabled] (cap: $N or none; resets day N at HH:MM UTC)

If the monthly spend budget was just enabled, append:

Monthly spend note: the "mo:" segment is Claude Code's local cost estimate for sessions
on this machine rendered from now on. It will not match the Anthropic console exactly;
use the console usage page for the authoritative figure.