[FEATURE] Display CLAUDE.md loading status at startup

Preflight Checklist

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

Problem Statement

Claude Code automatically loads CLAUDE.md files but provides no feedback when:

1. No CLAUDE.md file is found in the project
2. CLAUDE.md files are successfully loaded
3. Users accidentally name their context file incorrectly (e.g., AGENT.md, README.md)

This creates a silent failure mode where users think their project guidelines are being followed, but Claude Code has no context loaded. Users
discover this only after observing incorrect behavior or non-compliance with project-specific rules.

Proposed Solution

Add a startup notification system that displays the status of context file loading:

1. Success case: "✅ Loaded context from CLAUDE.md"
2. Missing case: "⚠️ No CLAUDE.md found in project. Claude Code will proceed without project-specific context."
3. Alternative file detected: "⚠️ Found AGENT.md but it won't be auto-loaded. Rename to CLAUDE.md for automatic loading."

This could be:

• Displayed in the first system message when starting a conversation
• Added to the output of the /memory command
• Shown in a dedicated status indicator in the CLI interface

Alternative Solutions

Alternative approaches that could work:

1. Verbose mode flag: Add a --verbose or --show-context flag to display loaded context files at startup
2. Status command: Add a /status or /context command to check loaded files at any time
3. Configuration warning: When .claude/settings.local.json exists but no CLAUDE.md is found, show a warning
4. Auto-detection of common mistakes: Detect files like AGENT.md, CLAUDE.txt, claude.md and suggest renaming

Priority

Medium - Would be very helpful

Feature Category

CLI commands and flags

Use Case Example

Scenario: A team creates an AGENT.md file (following emerging industry conventions) with critical project guidelines:

• "NEVER run build commands - user handles builds manually"
• "Always use Strings.isNullOrEmpty() for string checks"
• "No AI metadata in git commits"

Current behavior:

1. User starts Claude Code session
2. No indication that AGENT.md is not being loaded
3. Claude attempts to run sbt compile, violating project rules
4. User discovers context wasn't loaded, renames to CLAUDE.md
5. No confirmation the file is now loaded - user must test behavior again

Desired behavior:

1. User starts Claude Code session
2. Display: "⚠️ Found AGENT.md but not loaded. Rename to CLAUDE.md for automatic loading."
3. User renames file
4. Next session shows: "✅ Loaded context from CLAUDE.md"
5. User has confidence their guidelines are active

Additional Context

Related observations:

1. Documentation clarity: While the docs explain CLAUDE.md auto-loading, they don't emphasize that NO feedback is given when files are

loaded/missing

1. Industry evolution: The convention file naming is evolving (AGENT.md, .cursorrules, etc.). Claude Code's strict requirement for

"CLAUDE.md" makes it important to warn users when they use alternative names

1. Multi-level loading: Claude Code supports hierarchical CLAUDE.md files (~/.claude/, project root, subdirectories). A startup message

could show ALL loaded context files with their paths

1. Comparison to other tools: Many CLI tools display configuration status at startup (git shows branch, docker shows running containers,

etc.). This is a familiar UX pattern

1. /memory command limitation: The /memory command shows loaded context but requires the user to actively check it. Many users won't know to

do this

This feature would significantly improve the "pit of success" for Claude Code users by making context loading explicit and verifiable.