Auto rename
Au-delà de l'efficience, l'osmose. Plus que des scripts, ces agents Claude Code sont les membres fantômes de notre intellect : des extensions vitales plutôt que de simples exécutants.
npx -y skills add Pamacea/smite --skill auto-renameAssembled from the repository path, not quoted from the project. Check it against their README if it does not work.
One thing to look at
- 0 stars0 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.
What its author says it does
Copied from the file, not written here
Automatically runs on SessionStart, PostToolUse, and UserPromptSubmit hooks to rename Claude Code sessions using LLM analysis following "Action: Context" format (e.g., "Fix: login bug", "Add: user CRUD"). Generates intelligent, context-aware names that evolve as session progresses. Manual override via /rename command available. Integrates with session .jsonl storage. (user)
SKILL.md
4.9 KB, ~1.0k tokens by cl100k_base, as published. Nobody here has run it
Mission
Automatically rename Claude Code sessions with intelligent, context-aware names generated by AI analysis.
Usage Guidelines
Automatic Activation
The skill runs automatically on three key hooks:
| Hook | When it Triggers | What it Does |
|---|---|---|
| SessionStart | When a session begins | Sets initial name from first meaningful message |
| PostToolUse | After code operations (Write, Bash, git) | Updates name based on work performed |
| UserPromptSubmit | After significant conversation | Refines name as session evolves |
Smart Filtering
The system intelligently avoids unnecessary renames by:
- Skipping command/skill invocations: Only processes genuine user requests
- Filtering system messages: Ignores <command->, <objective>, <process> artifacts
- Rate limiting: Maximum 10 automatic renames prevents spam
- Quality threshold: Only renames when meaningful improvement detected
Manual Override
# Automatic (runs on hooks - no action needed)
# The system handles everything automatically
# Manual override when you want custom names
/rename "Build JWT authentication system"
# Auto-generated examples (for reference)
"Fix: login bug"
"Add: user CRUD API"
"Refactor: auth system"
"Debug: memory leak"
Core Workflow
- Input: Hook trigger (SessionStart, PostToolUse, UserPromptSubmit)
- Process:
- Intelligently skip command/skill invocations (system-generated messages)
- Analyze session context (first real user message, recent tools, project)
- Generate concise name following "Action: Context" format
- Update session .jsonl file with system message
- Output: Session renamed and visible in history/resume
Key Principles
- Automatic operation: Runs in background without manual intervention
- Intelligent naming: LLM analysis understands session context
- Command artifact filtering: Skips <command->, <objective>, <process> tags from skills
- Dynamic updates: Name evolves as session progresses
- Manual override:
/renamecommand available for custom names - User control: Always respects manual naming preferences
Name Format
Pattern: Action: Context
| Action Prefix | Usage | Example |
|---|---|---|
| Fix | Bug fixes | Fix: login bug |
| Add | New features | Add: user CRUD API |
| Update | Modifications | Update: database schema |
| Delete | Removals | Delete: deprecated code |
| Refactor | Restructuring | Refactor: auth system |
| Debug | Investigation | Debug: memory leak |
| Test | Testing | Test: payment flow |
| Docs | Documentation | Docs: API README |
| Config | Configuration | Config: env variables |
Integration
- SessionStart hook: Analyzes first message, sets initial name
- PostToolUse hook: Renames after Write/Bash/git operations
- UserPromptSubmit hook: Updates based on conversation evolution
- Manual command:
/renamefor custom or auto-generated names - Storage: System messages in .jsonl session files
Configuration
- maxSessionNameLength: 50 characters (default)
- maxRenamesPerSession: 10 automatic renames (default)
- renameTriggers: Control which hooks activate (sessionStart, postToolUse, userPromptSubmit)
- Config file:
config/settings.json
Error Handling
- Insufficient content: Falls back to "Working on code"
- LLM failure: Generic name, retry on next trigger
- Network issues: Graceful degradation to defaults
- File permissions: Report error, skip rename
- Max renames reached: Stop automatic renaming, manual still available
Design Principles
| Principle | Implementation | Benefit |
|---|---|---|
| Rate Limiting | Maximum 10 automatic renames per session | Prevents spam, maintains focus |
| Smart Filtering | Filters out <command->, <objective>, <process> tags | Clean, meaningful names |
| Specific Naming | Uses "Action: Context" format with clear prefixes | Instant session understanding |
| Non-Intrusive Operation | Skips system-generated and skill execution messages | No distraction during work |
| Contextual Evolution | Updates name based on conversation progress | Names stay relevant |
| User Sovereignty | Manual /rename command always available | Full control when needed |
Auto-generated from plugin.json - Last sync: 2025-01-22
What ships with it
Read from the repository
Just SKILL.md. No reference files, no scripts.