jpmschweitzerandClaude Opus 4.6
7158ae9cf5
Claude-native starter kit that bootstraps multi-agent team infrastructure for any project. Clone once, install as a global skill, run /kit-install in any project directory. Includes: - 3-tier profile system (minimal/standard/full: 3-12 agents) - 16 agent archetype templates with personality spectrum - 18 skill templates using domain-action naming convention - Stakeholder persona panel for workshops and PR reviews - SQLite ticketing DB with CLI tools (config-based DB paths) - Decision tracking, sprint lifecycle, workshop orchestration - Multi-git-host support (GitHub, Gitea, GitLab) - /kit-update skill for syncing with source repo evolution - Naming theme support for agent identity/flavor - Smoke tests for all three profile tiers Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
3.0 KiB
3.0 KiB
name, description, user-invocable, allowed-tools
| name | description | user-invocable | allowed-tools |
|---|---|---|---|
| debt-scan | Scan for technical debt and generate a debt registry. Use when the user says "tech debt", "debt scan", "code quality", or invokes /debt-scan. Identifies TODO/FIXME comments, lint warnings, complex functions, and other debt signals. | true | Bash, Read, Grep, Glob, Write, AskUserQuestion |
Technical Debt Scanner
Scan the codebase for technical debt signals and generate a prioritized debt registry.
Workflow
1. Detect Tech Stack and Run Linters
Identify the project's languages and run appropriate linters:
| Stack | Lint Command |
|---|---|
| Rust | cargo clippy -- -D warnings 2>&1 |
| JavaScript/TypeScript | npx eslint . --format json 2>/dev/null |
| Python | ruff check . or pylint --output-format=json **/*.py 2>/dev/null |
| GDScript | Godot headless check for SCRIPT ERROR |
Capture warning and error counts from each linter.
2. Scan for Debt Markers
Search the codebase for explicit debt signals:
- Action comments:
TODO,FIXME,HACK,WORKAROUND,TEMPORARY - Deprecation markers:
@deprecated,#[deprecated] - Lint suppressions:
#[allow(dead_code)],// eslint-disable,# noqa,@SuppressWarnings
Record the file, line number, and surrounding context for each match.
3. Complexity Analysis
Identify structural debt:
- Large files: files exceeding 500 lines
- Complex functions: high cyclomatic complexity (if tooling is available)
- Deep nesting: code with more than 4 levels of indentation
- Long functions: functions exceeding 100 lines
4. Cross-Reference with Tickets
Check whether existing tickets already address found debt:
- Match debt locations against ticket descriptions and referenced files
- Identify debt items with no corresponding ticket (untracked debt)
- Flag tickets that reference debt already resolved
5. Score and Prioritize
Rate each debt item on two axes:
- Impact: high (affects users or stability), medium (affects developer productivity), low (cosmetic or stylistic)
- Effort: quick fix (< 1 hour), medium refactor (1 day), large rework (multi-day)
Priority = impact weight / effort weight — surface high-impact, low-effort items first.
6. Output
Generate the debt registry:
## Tech Debt Registry — {date}
### Summary
Lint warnings: N | TODOs: N | Suppressions: N | Large files: N
### High Priority
| Location | Type | Description | Effort | Impact |
|----------|------|-------------|--------|--------|
### Medium Priority
| Location | Type | Description | Effort | Impact |
|----------|------|-------------|--------|--------|
### Low Priority
| Location | Type | Description | Effort | Impact |
|----------|------|-------------|--------|--------|
### Trends
{Comparison with previous scan if available — growing/shrinking debt}
7. Optional: Create Tickets
Ask the user (via AskUserQuestion) if they want to create tickets for the top-priority debt items. If yes, create tickets with appropriate priority and team assignment.