Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Use only when a developer explicitly asks to clean, prune, tidy, or organize DeepChat SDD documentation after implementation and validation. Scans docs/features, docs/issues, and docs/architecture; prefers multi-agent review when available; removes completed issue docs tied to closed GitHub issues, drops stale plan/tasks files from completed feature or architecture goals, and deletes obsolete feature or architecture docs.
.claude/skills/thinkinaixyz-deepchat-sdd-cleanup/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-08 | ✗→✓ | ▲ Improved | 9% | 0% |
| case-09 | ✗→✓ | ▲ Improved | 26% | 0% |
| case-15 | ✗→✓ | ▲ Improved | -28% | 0% |
| case-22 | ✗→✓ | ▲ Improved | 26% | 0% |
| case-16 | ✓→✓ | ＝ Same ✓ | -25% | 0% |
Run this skill only when the developer explicitly asks for SDD cleanup, documentation tidying, pruning, or removal of completed/stale SDD files. Do not run it as an automatic final step of ordinary feature, bug, architecture, or release work.
docs/spec-driven-dev.md, docs/README.md, and git status.docs/features, docs/issues, and docs/architecture with find or rg.docs/featuresdocs/issuesdocs/architectureplan.md and any legacy tasks.md; keep spec.mdonly when it still defines a maintained contract, regression guard, platform policy, or architecture decision.
implementation and validation evidence prove the bug no longer exists.
decision record.
describes a maintained boundary; otherwise update the spec.
leaving contradictory docs.
Use gh only when it is installed and authenticated. For linked issue docs, verify closure with gh issue view <number> --json state,url,title when possible. If gh is unavailable, do not delete solely because a GitHub link looks old.
plan.md steps or legacy tasks.[NEEDS CLARIFICATION].docs/README.md, docs/ARCHITECTURE.md, docs/FLOWS.md, or AGENTSinstructions unless the reference is updated in the same change.
rg -n "plan.md|tasks.md|docs/archives|NEEDS CLARIFICATION" docs AGENTS.md .agents/skillsand inspect any stale policy references.
git status --short.touched generated files require them.
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-16 | pass→pass | 12,655 | 5,484 | -57% | 1 | 1 | 0% | 2,064 | 1,540 | -25% | 0 | 0 | — |
case-17 | pass→pass | 11,399 | 2,682 | -76% | 1 | 1 | 0% | 1,666 | 1,030 | -38% | 0 | 0 | — |
case-01 | fail→fail | 2,912 | 5,968 | +105% | 1 | 1 | 0% | 294 | 890 | +203% | 0 | 0 | — |
case-02 | fail→fail | 2,283 | 5,958 | +161% | 1 | 1 | 0% | 315 | 1,064 | +238% | 0 | 0 | — |
case-03 | fail→fail | 12,289 | 5,603 | -54% | 1 | 1 | 0% | 1,849 | 922 | -50% | 0 | 0 | — |
case-04 | pass→pass | 11,129 | 4,181 | -62% | 1 | 1 | 0% | 1,599 | 1,210 | -24% | 0 | 0 | — |
case-05 | fail→fail | 6,701 | 6,254 | -7% | 1 | 1 | 0% | 366 | 965 | +164% | 0 | 0 | — |
case-06 | fail→fail | 22,911 | 6,214 | -73% | 1 | 1 | 0% | 3,703 | 814 | -78% | 0 | 0 | — |
case-07 | pass→pass | 7,533 | 4,319 | -43% | 1 | 1 | 0% | 1,263 | 1,384 | +10% | 0 | 0 | — |
case-08 | fail→pass | 7,259 | 3,381 | -53% | 1 | 1 | 0% | 1,072 | 1,164 | +9% | 0 | 0 | — |
case-09 | fail→pass | 7,650 | 4,991 | -35% | 1 | 1 | 0% | 1,158 | 1,457 | +26% | 0 | 0 | — |
case-10 | pass→pass | 7,276 | 5,268 | -28% | 1 | 1 | 0% | 1,129 | 1,623 | +44% | 0 | 0 | — |
case-11 | pass→pass | 4,596 | 2,840 | -38% | 1 | 1 | 0% | 702 | 1,062 | +51% | 0 | 0 | — |
case-12 | pass→pass | 10,007 | 4,300 | -57% | 1 | 1 | 0% | 1,457 | 1,365 | -6% | 0 | 0 | — |
case-13 | pass→pass | 10,127 | 4,780 | -53% | 1 | 1 | 0% | 1,525 | 1,436 | -6% | 0 | 0 | — |
case-14 | pass→pass | 8,711 | 6,650 | -24% | 1 | 1 | 0% | 1,238 | 1,701 | +37% | 0 | 0 | — |
case-15 | fail→pass | 11,720 | 3,637 | -69% | 1 | 1 | 0% | 1,777 | 1,274 | -28% | 0 | 0 | — |
case-18 | pass→pass | 10,823 | 4,594 | -58% | 1 | 1 | 0% | 1,546 | 1,368 | -12% | 0 | 0 | — |
case-19 | pass→pass | 7,947 | 3,638 | -54% | 1 | 1 | 0% | 1,229 | 1,303 | +6% | 0 | 0 | — |
case-20 | pass→pass | 9,921 | 6,127 | -38% | 1 | 1 | 0% | 1,388 | 1,584 | +14% | 0 | 0 | — |
case-21 | pass→pass | 6,708 | 3,537 | -47% | 1 | 1 | 0% | 1,035 | 1,228 | +19% | 0 | 0 | — |
case-22 | fail→pass | 5,010 | 2,618 | -48% | 1 | 1 | 0% | 767 | 964 | +26% | 0 | 0 | — |
DecimalAI ran this skill against gemini-3.6-flash twice over the same eval suite — once with the skill loaded and once without — and compared the two runs case by case. 22 cases were attempted, and 17 counted toward the lift figure. The other 5 produced results that are not comparable between the two arms, so they are excluded from the headline rather than averaged into it. The headline lift of +18 percentage points is the difference between those two pass rates over the 17 comparable cases.
Without the skill loaded, the model failed this case. With it loaded, the same prompt on the same model passed. This is one improved case from the latest verified run; every case, including any that regressed, is in the table above.
| Model | Method | Date | Lift |
|---|---|---|---|
| gemini-3.6-flash | verified | 8/9/2026 | +16% |
Other measured skills in the registry, with their headline benchmark lift.