.claude/skills/gap-analysis/SKILL.md
Find gaps, stubs, broken workflows, and incomplete features in code. Use before implementing features, for code audits, when checking TODO coverage, or when searching for buttons without handlers, forms without submit, links to nowhere, or partial CRUD implementations.
npx skillsauth add dmitryprg-ai/cursor-develop-autorules gap-analysisInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
Find implicit requirements and incomplete code. 90% of AI failures come from specification gaps — what was never stated but silently assumed.
List only what is explicitly stated. Quote the user's words.
| Category | Questions | Example Gaps | |----------|-----------|-------------| | Data | Formats? Null/empty? Boundaries? | max length, encoding, special chars | | UX | Loading state? Error state? Empty state? | skeleton, retry button, placeholder | | Integration | What else is affected? Side effects? | cache invalidation, events | | Security | Access rights? Validation? | authorization, input sanitization | | Performance | Data volume? Pagination? | batch limits, timeouts | | Errors | What if it fails? Retry? | graceful degradation, error messages |
Ask: "A week passed, the feature broke. Why?"
Common answers: missed edge case, uncovered error scenario, forgot state handling, didn't check access rights.
For HIGH gaps — ask the user, don't assume. Assumptions about critical gaps are the most dangerous kind of technical debt.
Run automated detection:
bash ${CLAUDE_SKILL_DIR}/scripts/scan-gaps.sh
Manual checks:
| Check | Pattern | Gap if |
|-------|---------|--------|
| Buttons without actions | onClick={() => {}} | Button does nothing |
| Forms without submit | onSubmit missing | Form doesn't send data |
| Links to nowhere | href="#" | Link leads nowhere |
| Loading without timeout | isLoading without error | May hang forever |
| Empty catches | catch {} | Error swallowed |
| Partial CRUD | Create but no Update/Delete | Incomplete |
| Severity | Criteria | Action | |----------|----------|--------| | HIGH | Feature doesn't work without it | Blocker — fix before implementation | | MED | Works but bad UX/quality | Fix during implementation | | LOW | Nice-to-have | Backlog or ask user |
Before creating anything new:
rg "X" — maybe it already exists?rg "TODO.*X" — maybe there's a stub?If a stub exists, complete it instead of creating a duplicate.
development
Scan codebase for technical debt and fix safely with TDD. Use to find oversized files, duplicated code, code smells, and refactor safely. Workflow - SCAN, TEST CASES, REFACTOR, VERIFY. Keywords - techdebt, tech debt, duplicates, code quality audit.
development
Test-Driven Development workflow with strict Red-Green-Refactor cycle. Use when developing features with TDD, writing tests before code, or when test-driven approach is needed. MANDATORY order - test cases table BEFORE code, failing tests BEFORE implementation.
testing
Review work session quality and capture improvements. Use at end of session, after large tasks, after series of errors, or when user asks for session review, retrospective, lessons learned. Records improvements to backlog.
data-ai
Analyze data, investigate datasets, work with CSV/parquet/pandas/dataframes. Use when analyzing data, exploring datasets, running experiments, or when user mentions data, analysis, parquet, csv, pandas, dataframe, statistics, investigation.