メニュー
Logs, categorizes, and prioritizes tech debt as it's incurred. Not a shame exercise -- a managed backlog with pay-down window recommendations and priority scoring.
Codex または Claude でインストール この Prompt をコピーして Codex、Claude、または他のアシスタントに貼り付けると、Skill ページを確認してインストールできます。
SOC 職業分類に基づく
Master orchestrator for the Solo Dev Suite — a collection of lifecycle skills (scope guardian, sprint planner, pricing architect, launch readiness, ADR generator, security audit, integration mapper, auto-docs, tech debt register, testing strategy) that share a central Project Profile. Triggers on requests like "solo dev suite", "run the suite on {project}", "set up a project profile", "what suite skills apply here", "show my projects", or any direct invocation of a child skill by name when no profile exists yet. Not for one-off tasks with no ongoing project — those stay as standalone skills.
Generates and maintains Architecture Decision Records (ADRs) in the Michael Nygard format -- numbered, sequentially stored Markdown docs capturing context, options considered, decision, and consequences. Triggers on "generate an ADR", "record this decision", "log architecture decision", "why did I choose X", "add an ADR", or "list my ADRs". Part of the solo-dev-suite -- loads the project profile via the orchestrator. Not for runtime config changes (those are just code) or meeting notes (use a wiki).
Generates and maintains project documentation (README, SETUP, ARCHITECTURE, CHANGELOG) by pulling from the profile and sibling sidecars. Re-runnable at every milestone so docs stay current without being a full-time job.
Scans a codebase for local-to-cloud migration blockers -- hardcoded paths, LAN URLs, dev-only bypasses, missing deploy configs -- and produces a scored readiness report. Triggers on "deploy readiness", "ready to deploy", "can I host this", "cloud migration check", "scan for local stuff", "deployment blockers". Part of the solo-dev-suite -- loads the project profile via the orchestrator. Unlike security-audit (a checklist), this skill actually reads code.
Generates a customized design-exploration loop prompt for an existing project by discovering its stack, domain, and aesthetic, then tailoring references and constraints accordingly. Triggers on requests like "design loop on {project}", "refine the UI of this project", "explore design directions for {project}", "make this look more polished", "visual refresh", "modernize the UI", or "UI audit". Not for greenfield design, CSS debugging, or single-component tweaks — requires a target directory with an existing UI surface to analyze.
Comprehensive feature-enhancement discovery skill. Points at any codebase — the current Claude Code workspace, a local source folder, or a Forgejo/Git repo URL — scans the full stack, inventories existing features, surfaces incomplete/stub work, then researches comparable apps and platforms to generate a prioritized list of value-add enhancements. Outputs a structured FEATURE_ENHANCEMENTS.md report. Trigger whenever the user says: "feature enhancement", "what features are we missing", "analyze my codebase for improvements", "compare to competitors", "what should I add next", "audit my app", "run the enhancement skill", "scan for TODOs and features", "what else should this do", or points at a project and asks for improvement ideas. Also triggers when user wants to know how their app stacks up against similar tools or what features competitors have that they don't.
| name | tech-debt-register |
| version | 1.0.0 |
| description | Logs, categorizes, and prioritizes tech debt as it's incurred. Not a shame exercise -- a managed backlog with pay-down window recommendations and priority scoring. |
log.Every item is tagged with one or more: design, code, infra, docs, dependencies, security, ui-ux, testing, performance.
priority = (impact_rank * urgency_rank) / effort_rank
Where impact: low=1, medium=2, high=4, critical=8; urgency: never=0, when-it-bites=1, post-launch-90d=2, post-launch-30d=3, pre-launch=4, now=5; effort: S=1, M=2, L=3, XL=4.
Higher score = pay down first. list --recommend sorts by this.
Reopening moves any status back to open.
Log a new debt item. Sidecar is auto-created on first log.
echo '{"title":"No tests on builder","description":"Shipped without coverage","categories":["testing"],"impact":"high","effort":"L","urgency_window":"post-launch-30d"}' \
| python scripts/debt_tool.py log <slug> --from-stdin
Required: title, description. Defaults: impact=medium, effort=M, urgency=when-it-bites, categories=["code"].
List items with optional filters and recommendation sorting.
python scripts/debt_tool.py list <slug>
python scripts/debt_tool.py list <slug> --recommend
python scripts/debt_tool.py list <slug> --status paid-down
python scripts/debt_tool.py list <slug> --category testing
Show full details of a single item.
python scripts/debt_tool.py show <slug> --id TD01
Mark as paid-down with resolution notes.
python scripts/debt_tool.py resolve <slug> --id TD01 --resolution-notes "Added unit tests"
Mark as won't-fix with required rationale.
python scripts/debt_tool.py accept <slug> --id TD01 --reason "Not worth the effort for this feature"
Move resolved/accepted back to open.
python scripts/debt_tool.py reopen <slug> --id TD01
Generate TECH_DEBT.md with summary, recommendations, open/resolved/accepted sections.
python scripts/debt_tool.py render <slug>
Remove the sidecar. No generated files to clean up.
python scripts/debt_tool.py delete <slug> --yes
| File | Purpose |
|---|---|
scripts/debt_tool.py | CLI with all subcommands |
templates/techdebt.schema.json | Sidecar JSON Schema |
templates/TECH_DEBT.md.tmpl | Rendered doc template |
# 1. Log two items
echo '{"title":"No tests on workout builder","description":"Shipped without tests","categories":["testing"],"impact":"high","effort":"L","urgency_window":"post-launch-30d"}' \
| python scripts/debt_tool.py log my-project --from-stdin
echo '{"title":"Stripe webhook not idempotent","description":"Duplicate deliveries cause duplicate charges","categories":["code","infra"],"impact":"critical","effort":"S","urgency_window":"pre-launch"}' \
| python scripts/debt_tool.py log my-project --from-stdin
# 2. List + recommend (TD02 should rank first: critical/S/pre-launch)
python scripts/debt_tool.py list my-project --recommend
# 3. Resolve TD02
python scripts/debt_tool.py resolve my-project --id TD02 --resolution-notes "Added event_id tracking"
# 4. List open only (should show TD01 only)
python scripts/debt_tool.py list my-project --status open
# 5. Render
python scripts/debt_tool.py render my-project
# 6. Delete
python scripts/debt_tool.py delete my-project --yes