aggregate-logs
Generate LEARNINGS.md from skill execution logs over a configurable time window.
Audit and sync plugin.json registrations with actual disk contents. Detects missing or stale skills, commands, agents, hooks.
> /plugin marketplace add athola/claude-night-marketHow it fires
How this command gets triggered: by you, by Claude, or both.
/update-pluginsContext preview
What this command does when you run it.
Audit and sync plugin.json registrations with actual disk contents. Detects missing or stale skills, commands, agents, hooks.
description: Audit and sync plugin.json registrations with actual disk contents. Detects missing or stale skills, commands, agents, hooks. usage: /update-plugins [plugin-name] [--dry-run] [--fix]
Audit plugin.json files against actual disk contents and fix registration gaps.
1. Scans each plugin directory for commands, skills, agents, hooks on disk 2. Compares with plugin.json registrations 3. Reports discrepancies: missing registrations, stale entries 4. Optionally fixes by updating plugin.json files with proper sorting
Runs `validate_plugin.py` on each affected plugin to verify:
python3 plugins/abstract/scripts/validate_plugin.py plugins/<plugin>
Checks all hooks for outdated patterns against the Claude Code SDK spec.
1. **Static check** (always): Runs `check_hook_modernization.py` to detect invalid decision values, deprecated fields, and missing error handling 2. **Tome research** (default, skip with `--skip-research`): Dispatches `tome:code-search` to verify patterns against current community practices
# Static check only python3 scripts/check_hook_modernization.py # JSON output for programmatic use python3 scripts/check_hook_modernization.py --json
See `update-plugins/modules/phase1c-modernization.md` for pattern reference and fix instructions.
Triggers `/plugin-review --tier branch` on affected plugins. Runs quick quality gates (test, lint, typecheck, registration, structure validation) on changed plugins and side-effect checks on related plugins.
These modules contain full procedural details: format templates, commands, TodoWrite naming conventions, and auto-issue creation procedures.
skill-review and performance analysis
recursive quality validation
queue promotion checks
Execute the Python script with the provided arguments:
# Audit all plugins (dry run - show discrepancies only) python3 plugins/sanctum/scripts/update_plugin_registrations.py --dry-run # Audit specific plugin python3 plugins/sanctum/scripts/update_plugin_registrations.py parseltongue --dry-run # Fix all plugins (update plugin.json files) python3 plugins/sanctum/scripts/update_plugin_registrations.py --fix # Fix specific plugin python3 plugins/sanctum/scripts/update_plugin_registrations.py abstract --fix
| Flag | Purpose | |------|---------| | `--dry-run` | Show discrepancies without fixing (default) | | `--fix` | Update plugin.json files | | `--skip-modernization` | Skip Phase 1c modernization audit | | `--skip-research` | Skip tome research (keep static check) | | `--skip-meta-eval` | Skip meta-evaluation check | | `--skip-queue` | Skip knowledge queue check | | `--no-auto-issues` | Don't auto-create GitHub issues |
This command requires the sanctum plugin's Python scripts. Before running, verify the script exists:
# Check if running in night-market repo if [[ -f "plugins/sanctum/scripts/update_plugin_registrations.py" ]]; then echo "Sanctum scripts available - command will work" else echo "ERROR: This command only works in the claude-night-market repository" echo "The update_plugin_registrations.py script is not available in this project." exit 1 fi
**For non-night-market projects**: This command is night-market-specific. Use `claude plugin update <plugin>@<marketplace>` to update individual plugins.
__pycache__, test files, __init__.py
| Type | Meaning | Action | |------|---------|--------| | Missing | File on disk, not in plugin.json | Add registration | | Stale | In plugin.json, not on disk | Remove or investigate | | Path mismatch | Wrong path format | Correct path |
degradation early
stability
Phases 1-4 create a continuous feedback loop: registration audit surfaces gaps, performance analysis identifies degradation, meta-evaluation enforces quality, and queue checks prevent knowledge loss. See the phase modules for details.
This command complements:
Phase 2)
changes
changes
A plugin marketplace for Claude Code. Install only the plugins you need to run git workflows, code review, spec-driven development, and autonomous agents from inside your Claude Code session.
Generate LEARNINGS.md from skill execution logs over a configurable time window.
Analyze skill file complexity metrics and generate modularization recommendations for splitting or progressive loading.
Scaffold new Claude Code skills with brainstorming, TDD methodology, and proper frontmatter and module structure.