/tech-debt

name: tech-debt
type: workflow
description: "Scans the codebase for technical debt indicators, categorizes and prioritizes findings, and maintains a debt register with repayment recommendations. Use when assessing codebase health, planning refactoring, or when the user mentions technical debt or code quality."
argument-hint: "[scan|add|prioritize|report]"
user-invocable: true
allowed-tools: Read, Glob, Grep, Write
context: fork
agent: technical-director
effort: 3
when_to_use: "Use when assessing codebase health, planning a refactoring sprint, or when the user mentions technical debt, code quality, cleanup, or wants to prioritize remediation work."

When this skill is invoked:

1. **Parse the subcommand** from the argument:

• `scan` — Scan the codebase for tech debt indicators
• `add` — Add a new tech debt entry manually
• `prioritize` — Re-prioritize the existing debt register
• `report` — Generate a summary report of current debt status

2. **For `scan`**:

• Search the codebase for debt indicators:
• `TODO` comments (count and categorize)
• `FIXME` comments (these are bugs disguised as debt)
• `HACK` comments (workarounds that need proper solutions)
• `@deprecated` markers
• Duplicated code blocks (similar patterns in multiple files)
• Files over 500 lines (potential god objects)
• Functions over 50 lines (potential complexity)
• Shallow wrappers or pass-through abstractions that fail the deletion test
• Categorize each finding:
• **Architecture Debt**: Wrong abstractions, missing patterns, coupling issues, shallow wrappers, failed deletion-test seams
• **Code Quality Debt**: Duplication, complexity, naming, missing types
• **Test Debt**: Missing tests, flaky tests, untested edge cases
• **Documentation Debt**: Missing docs, outdated docs, undocumented APIs
• **Dependency Debt**: Outdated packages, deprecated APIs, version conflicts
• **Performance Debt**: Known slow paths, unoptimized queries, memory issues
• Update the debt register at `docs/tech-debt-register.md`

For shallow-wrapper findings, apply this test before filing debt:

• If deleting the abstraction would mostly simplify the code and would not

force multiple callers to re-learn meaningful behavior, log it as **Architecture Debt**.

• If deleting it would re-spread invariants, branching, retries, validation,

or ordering logic across callers, it is probably earning its keep; do not file debt just because it is small.

Record shallow-wrapper debt explicitly in the description, for example:

• "Pass-through adapter around `OrderRepository` fails deletion test"
• "Thin mapper module mirrors caller complexity without adding leverage"

3. **For `add`**:

• Prompt for: description, category, affected files, estimated fix effort, impact if left unfixed
• Append to the debt register

4. **For `prioritize`**:

• Read the debt register
• Score each item by: `(impact_if_unfixed * frequency_of_encounter) / fix_effort`
• Re-sort the register by priority score
• Recommend which items to include in the next sprint

5. **For `report`**:

• Read the debt register
• Generate summary statistics:
• Total items by category
• Total estimated fix effort
• Items added vs resolved since last report
• Trending direction (growing / stable / shrinking)
• Flag any items that have been in the register for more than 3 sprints
• Output the report

Debt Register Format

## Technical Debt Register
Last updated: [Date]
Total items: [N] | Estimated total effort: [T-shirt sizes summed]

| ID | Category | Description | Files | Effort | Impact | Priority | Added | Sprint |
|----|----------|-------------|-------|--------|--------|----------|-------|--------|
| TD-001 | [Cat] | [Description] | [files] | [S/M/L/XL] | [Low/Med/High/Critical] | [Score] | [Date] | [Sprint to fix or "Backlog"] |

Rules

• Tech debt is not inherently bad — it is a tool. The register tracks conscious decisions.
• Every debt entry must explain WHY it was accepted (deadline, prototype, missing info)
• "Scan" should run at least once per sprint to catch new debt
• Items older than 3 sprints without action should either be fixed or consciously accepted with a documented reason
• Do not file every small abstraction as debt. File only wrappers/modules that

are shallow enough to fail the deletion test.

Protocol

• **Question**: Reads mode from argument (`scan` / `add` / `prioritize` / `report`); `add` mode prompts for debt details
• **Options**: Skip — mode drives execution path
• **Decision**: `add` mode — user provides description, category, effort estimate, and justification (why accepted)
• **Draft**: Changes shown in conversation before updating register
• **Approval**: "May I update `docs/technical/tech-debt.md`?"

Output

Deliver exactly:

• **`scan`**: debt register updated at `docs/technical/tech-debt.md` + count of new items found by category
• **`add`**: new TD-[NNN] entry appended to register with effort, impact, and WHY it was accepted
• **`prioritize`**: ranked list of top 5 items by impact/effort score with recommended sprint to fix
• **`report`**: summary table — total items, effort distribution (S/M/L/XL), critical items count