Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Deprecated — forwards to bmad-deep-recon (domain type)
.claude/skills/bmad-code-org-bmad-domain-research/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | -33% | 0% |
| case-03 | ✗→✓ | ▲ Improved | -44% | 0% |
| case-08 | ✗→✓ | ▲ Improved | -1% | 0% |
| case-09 | ✗→✓ | ▲ Improved | -21% | 0% |
| case-10 | ✗→✓ | ▲ Improved | 29% | 0% |
This skill was consolidated into bmad-deep-recon. It is retained as a thin compatibility shim so existing invocations by name and _bmad/custom/bmad-domain-research.toml override files keep working. New work should invoke bmad-deep-recon directly — it drafts deep-research prompts for outside tools, processes finished reports into downstream-ready summaries, and runs research directly, across market, domain, technical, competitive, user-voice, and academic-lit types (plus a select shape for choose-between decisions and custom types).
uv run {project-root}/_bmad/scripts/resolve_customization.py --skill {skill-root} --project-root {project-root} --key workflow. This picks up any {project-root}/_bmad/custom/bmad-domain-research.toml and bmad-domain-research.user.toml overrides for the legacy fields (activation_steps_prepend, activation_steps_append, persistent_facts, on_complete).bmad-domain-research is deprecated and forwards to bmad-deep-recon with the domain type. To silence this notice and access the full new surface (draft/process/run modes, research types, verification levels, HTML briefing, handoffs), migrate _bmad/custom/bmad-domain-research.toml to _bmad/custom/bmad-deep-recon.toml and invoke bmad-deep-recon directly.bmad-deep-recon with: research type domain (skip its type inference), the four legacy fields above as pre-resolved values, and the user's original input verbatim. bmad-deep-recon takes the workflow from here — do not execute any further steps in this shim.| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→pass | 51,690 | 31,457 | -39% | 1 | 1 | 0% | 6,829 | 4,588 | -33% | 0 | 0 | — |
case-02 | fail→fail | 68,403 | 29,079 | -57% | 1 | 1 | 0% | 7,626 | 4,756 | -38% | 0 | 0 | — |
case-03 | fail→pass | 174,237 | 73,286 | -58% | 1 | 1 | 0% | 8,226 | 4,647 | -44% | 0 | 0 | — |
case-04 | pass→pass | 4,602 | 7,599 | +65% | 1 | 1 | 0% | 712 | 1,490 | +109% | 0 | 0 | — |
case-05 | pass→pass | 10,572 | 19,002 | +80% | 1 | 1 | 0% | 1,974 | 2,347 | +19% | 0 | 0 | — |
case-06 | pass→pass | 5,824 | 4,689 | -19% | 1 | 1 | 0% | 1,021 | 1,132 | +11% | 0 | 0 | — |
case-07 | fail→fail | 35,072 | 37,930 | +8% | 1 | 1 | 0% | 3,798 | 3,975 | +5% | 0 | 0 | — |
case-08 | fail→pass | 24,308 | 19,759 | -19% | 1 | 1 | 0% | 3,727 | 3,679 | -1% | 0 | 0 | — |
case-09 | fail→pass | 77,787 | 41,793 | -46% | 1 | 1 | 0% | 8,697 | 6,878 | -21% | 0 | 0 | — |
case-10 | fail→pass | 16,827 | 15,945 | -5% | 1 | 1 | 0% | 2,286 | 2,942 | +29% | 0 | 0 | — |
case-11 | fail→fail | 30,965 | 33,808 | +9% | 1 | 1 | 0% | 2,832 | 4,834 | +71% | 0 | 0 | — |
case-12 | fail→fail | 18,108 | 21,010 | +16% | 1 | 1 | 0% | 2,579 | 3,506 | +36% | 0 | 0 | — |
case-13 | fail→fail | 30,214 | 25,570 | -15% | 1 | 1 | 0% | 4,385 | 3,679 | -16% | 0 | 0 | — |
case-14 | fail→fail | 36,444 | 42,557 | +17% | 1 | 1 | 0% | 5,129 | 3,637 | -29% | 0 | 0 | — |
case-15 | fail→fail | 27,677 | 60,204 | +118% | 1 | 1 | 0% | 4,157 | 4,691 | +13% | 0 | 0 | — |
case-16 | pass→pass | 16,091 | 18,197 | +13% | 1 | 1 | 0% | 2,890 | 2,698 | -7% | 0 | 0 | — |
case-17 | fail→pass | 15,585 | 15,364 | -1% | 1 | 1 | 0% | 2,354 | 2,777 | +18% | 0 | 0 | — |
case-18 | pass→fail | 67,479 | 25,108 | -63% | 1 | 1 | 0% | 3,676 | 4,153 | +13% | 0 | 0 | — |
case-19 | fail→pass | 13,376 | 19,930 | +49% | 1 | 1 | 0% | 2,051 | 3,766 | +84% | 0 | 0 | — |
case-20 | fail→pass | 21,800 | 21,204 | -3% | 1 | 1 | 0% | 3,251 | 3,484 | +7% | 0 | 0 | — |
case-21 | fail→pass | 19,653 | 26,693 | +36% | 1 | 1 | 0% | 2,979 | 3,685 | +24% | 0 | 0 | — |
case-22 | fail→pass | 30,807 | 64,722 | +110% | 1 | 1 | 0% | 4,603 | 4,932 | +7% | 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. The headline lift of +41 percentage points is the difference between those two pass rates over the 22 comparable cases. 1 case got worse with the skill loaded, and it is included in that figure.
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.
Other measured skills in the registry, with their headline benchmark lift.