claude-code - 💡(How to fix) Fix feat: support dynamic statusMessage interpolation for hooks [1 comments, 2 participants]

Preflight Checklist

• I have searched existing requests and this feature hasn't been requested yet
• This is a single feature request (not multiple features)

Problem Statement

feat: support dynamic statusMessage interpolation for hooks

Problem

Hook statusMessage in .claude/settings.json is a static string. When hooks fire on tool calls, the activity log shows repeated generic labels like "After tool use: Edit" with no indication of which file is being processed or what the hook is doing.

In a typical session with 30+ tool calls, this produces 60-100 identical entries that provide no useful feedback to the user.

Current behavior

{
  "matcher": "Read|Write|Edit|Agent",
  "hooks": [{
    "type": "command",
    "command": "/path/to/hook.sh",
    "statusMessage": "Processing..."
  }]
}

Activity log shows:

- After tool use: Edit    → "Processing..."
- After tool use: Edit    → "Processing..."
- After tool use: Edit    → "Processing..."

No way to tell which file or what action each entry corresponds to.

Desired behavior

Allow statusMessage to interpolate variables from the hook context:

{
  "statusMessage": "Processing: {{tool_input.file_path}}"
}

Activity log would then show:

- After tool use: Edit    → "Processing: src/Controller/VehicleController.php"
- After tool use: Edit    → "Processing: .claude/settings.json"
- After tool use: Edit    → "Processing: backend/config/routes.yaml"

Proposed API

Support {{variable}} interpolation in statusMessage with access to the same context available to the hook script via stdin:

Fallback: if the variable is not present in the context, replace with empty string (so "{{tool_input.file_path}}" on a Bash call just shows blank, not a literal {{...}}).

Workaround attempted

We consolidated 3 PostToolUse hooks into 1 dispatcher to reduce the number of entries (3→1 per tool call). This helps but doesn't solve the readability problem — each entry still shows a generic static message.

Environment

• Claude Code web (claude.ai/code)
• Hooks: PreToolUse + PostToolUse with matchers for Edit|Write|Read|Bash
• Typical session: 30-50 tool calls → 60-100 hook activity entries

Proposed Solution

Allow statusMessage to support {{variable}} interpolation using the same context object that hooks receive via stdin.

Configuration:

{ "statusMessage": "Processing: {{tool_input.file_path}}" }

Available variables:

Variable Example {{tool_name}} Edit {{tool_input.file_path}} src/Entity/Vehicle.php {{tool_input.command}} git push origin main Fallback: If a variable is not present in the context, replace with empty string.

Alternative Solutions

Consolidated 3 PostToolUse hooks into 1 dispatcher script to reduce the number of UI entries from 3 to 1 per tool call. This reduces noise by 50% but each entry still shows a static generic message — the user still can't tell which file triggered which hook.

Priority

Medium - Would be very helpful

Feature Category

Configuration and settings

Use Case Example

I'm building a Symfony app with 20+ workflow hooks (pre-edit gates, auto-evidence tracking, status line generation) I use Claude Code web (claude.ai/code) and the activity log shows 60-100 entries per session Every entry says "After tool use: Edit" with statusMessage "Processing..." — all identical With dynamic interpolation, I'd see "Processing: src/Controller/VehicleController.php" and instantly know what each hook invocation corresponds to This would save me from expanding each entry individually to understand what happened

Additional Context

No response