audit-statuslines

Audit Status Lines Command

Audit Claude Code status line configurations for quality and cross-platform compatibility.

Initialization

Before auditing, initialize the environment:

1. Get the current UTC date for audit timestamps.
2. Capture the project root path for subagent communication.
3. Ensure the temp directory (.claude/temp/) exists.
4. Clean up any stale audit files if the user confirms.

The status-line-customization skill provides authoritative validation guidance (auto-loaded when this command runs).

What Gets Audited

• Script structure and execution
• JSON input handling
• Terminal color code usage
• Cross-platform compatibility (Windows, macOS, Linux)
• Helper function availability

Command Arguments

Argument	Description
(none)	Audit all discoverable status line scripts
--force	Audit regardless of modification status
--skip-validation	Skip finding validation (faster, but may include false positives)

Step 1: Discover Status Line Configurations

Check for statusLine settings in both project and user settings files.

For project settings, read .claude/settings.json and extract any statusLine value.

For user settings, detect the platform first: on Unix-like systems check ~/.claude/settings.json, on Windows check %USERPROFILE%\.claude\settings.json. Then extract any statusLine value.

Each statusLine value points to a script file. Build a list of discovered scripts with their scope (project or user) and full path.

If no custom status lines are configured, report this and provide guidance on how to create one.

Step 2: Parse Arguments

Check if the --force flag is present in the command arguments. Build the audit queue based on discovered scripts and the force flag.

Step 3: Present Audit Plan

Display audit mode (SMART or FORCE), scripts discovered, and list each with scope and last modified date.

Step 4: Execute Audits

For each script, spawn the statusline-auditor subagent with the following context:

• Scope (project or user)
• Full path to the script file
• Settings file reference
• Last audit date or "Never audited"
• Current audit date
• Project root path

Run subagents in parallel when multiple scripts exist.

Subagents write findings to .claude/temp/ as both JSON (for recovery/aggregation) and markdown (for human review). The main conversation thread collects results and updates audit logs using its Write/Edit tools.

Step 4.5: Validate Findings

Unless --skip-validation flag is present:

1. Spawn the audit-finding-validator agent with:
   • project_root: The captured project root path
   • audit_type: "statusline"
   • audit_files: List of .claude/temp/audit-*-statusline-*.json file paths
2. Wait for validation to complete
3. Read updated JSON files with validation results
4. Filter out FALSE_POSITIVE findings completely before aggregation
5. Note: Filtered findings are logged to .claude/temp/audit-filtered-findings.json

If --skip-validation flag is present:

• Skip validation phase entirely (current speed preserved)
• Present all findings without filtering
• Note in summary: "Validation: Skipped"

Step 5: Final Summary

Report total scripts audited, results by scope, and details table. List cross-platform or JSON handling issues with remediation steps.

Include validation statistics (if validation was performed):

• Validation performed: Yes/No
• Findings validated: X
• False positives filtered: Y
• Verified findings: Z
• Unverified findings: W

Important Notes

Script Requirements

Status line scripts must accept JSON input via stdin, parse it correctly, output formatted terminal text, and work across Windows (Git Bash/PowerShell), macOS, and Linux.

Cross-Platform Paths

Platform	Settings Location
Unix	~/.claude/settings.json
Windows	%USERPROFILE%\.claude\settings.json

Terminal Color Codes

Use ANSI escape codes correctly with fallbacks for terminals without color support.

Audit Log Location

All audit results are written to .claude/audit/statuslines.md.

Use /audit-log statuslines to view current audit status.

Example Usage

Example 1: Audit All Status Lines

User: /audit-statuslines

Claude: Discovering status line configurations...

## Audit Plan
**Mode**: SMART
**Scripts discovered**: 2

1. [project] .claude/statusline.sh
2. [user] ~/scripts/my-statusline.py

[Spawns statusline-auditor subagents]

## Audit Complete
| Scope | Script | Result | Score |
| --- | --- | --- | --- |
| project | statusline.sh | PASS | 100/100 |
| user | my-statusline.py | PASS | 98/100 |

Example 2: Force Audit

User: /audit-statuslines --force

Claude: Auditing all scripts (force mode)...