skills/mine.gap-close/SKILL.md
Use when the user says: 'close gaps in this design', 'fill gaps in the spec', 'lightweight design review', 'gap-close this doc', 'completeness review', or wants to verify a doc has all required content before implementation.
npx skillsauth add NodeJSmith/Claudefiles mine.gap-closeInstall 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.
Completeness review for design docs, briefs, task files, and general-purpose docs. Finds missing required content (not flawed content), fills gaps conversationally via Edit, and signs off. Not a substitute for /mine.challenge (quality of existing content).
$ARGUMENTS — file path, feature directory, short description, or empty (will ask).
Parse $ARGUMENTS. If a file path: read directly. If a directory: look for design.md, brief.md, or first .md. If empty:
AskUserQuestion:
question: "Which artifact should I review?"
header: "Target"
options:
- label: "design.md in current feature"
description: "Use the design.md in the current or most recent feature directory"
- label: "I'll type the path"
description: "Specify a file path or description"
Read the artifact. Detect type:
**Status:** draft/approved AND ## Problem → design doc## Key Decisions OR ## Scope Boundaries → brief## Deliverables OR first heading starts with # WP → work packageLoad the matching checklist from ${CLAUDE_HOME:-~/.claude}/skills/mine.gap-close/REFERENCE.md. When type is ambiguous, confirm before surveying.
Evaluate each checklist item against the artifact. Mark each PASS, GAP (severity: Blocker / Should-address / Nice-to-have), or N/A. Example: DD-02 GAP [Blocker] Goals lack measurable success metrics.
Present gap counts by severity and one-line descriptions, then:
AskUserQuestion:
question: "How to work through gaps?"
header: "Scope"
options:
- label: "All gaps"
description: "Walk through every gap in severity order"
- label: "Blockers only"
description: "Skip Should-address and Nice-to-have"
- label: "Take the list"
description: "Print gap list and exit — I'll handle manually"
Iterate gaps: Blocker → Should-address → Nice-to-have. For each gap:
<!-- TODO: review -->.Every 3-5 gaps, checkpoint via AskUserQuestion (header: "Checkpoint") with options: "Continue", "Pause — stop here", "Skip remaining — jump to sign-off".
Re-run the checklist on the updated artifact.
If all Blockers resolved:
AskUserQuestion:
question: "Artifact is complete."
header: "Sign-off"
options:
- label: "Approve"
description: "Mark ready — update status field if present"
- label: "Run full challenge"
description: "Invoke /mine.challenge for deeper critique"
- label: "Save and stop"
description: "Leave as-is, no status change"
If Blockers remain: report count, then AskUserQuestion (header: "Blockers") with options: "Continue — keep filling" or "Save and stop".
On "Approve": if the detected type is design doc and the artifact contains **Status:** draft, update to approved; otherwise leave any existing status unchanged. On "Run full challenge": invoke /mine.challenge <artifact-path>, loop back to sign-off after. On "Save and stop": confirm with "Saved. Resume with /mine.gap-close <path> later."
development
Use when the user says: 'create an issue', 'file an issue', 'open an issue', 'write an issue', 'new issue for this'. Codebase-aware issue creation — investigates the code to produce well-structured issues with acceptance criteria, affected areas, and enough detail for automated triage.
development
Use when the user says: 'triage issues', 'classify issues by complexity', 'assess issue complexity', 'find quick wins', 'which issues are small', 'batch issue assessment'. Batch codebase-aware issue triage — parallel Haiku subagents assess actual complexity and effort by reading the code, not just titles.
development
Use when the user says: "review my changes", "run the reviewers", "code and integration review", "readability review", "maintainability review", "sniff test this", "WTF check", "code smells", "is this code any good", "fresh eyes on this branch", "review this directory", "check this module". Dispatches three parallel reviewers — code, integration, and a readability pass — and consolidates findings into one prioritized report.
development
Use when the user says: "clean code check", "style review", "LLM smell check", "code hygiene", "nitpick this", "style check", "find style sins", "nitpicker review", "anal retentive review", "exhaustive style review", "no-filter style report". Dispatches three parallel stylistic checkers — llm-checker (training-bias patterns), lazy-checker (deferred debt), and nitpicker (style hygiene) — and consolidates findings into a report organized by checker with a Summary section for orchestration consumption.