Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Deprecated — forwards to bmad-prd (create intent)
.claude/skills/bmad-code-org-bmad-create-prd/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | -17% | 0% |
| case-02 | ✗→✓ | ▲ Improved | 69% | 0% |
| case-04 | ✗→✓ | ▲ Improved | 6% | 0% |
| case-07 | ✗→✓ | ▲ Improved | -11% | 0% |
| case-09 | ✗→✓ | ▲ Improved | -10% | 0% |
This skill was consolidated into bmad-prd. It is retained as a thin compatibility shim so existing invocations by name and _bmad/custom/bmad-create-prd.toml override files keep working. New work should invoke bmad-prd directly — it detects create / update / validate intent from the conversation.
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-create-prd.toml and bmad-create-prd.user.toml overrides for the legacy fields (activation_steps_prepend, activation_steps_append, persistent_facts, on_complete).{project-root}/_bmad/bmm/config.yaml (and config.user.yaml if present) to resolve {user_name} and {communication_language}.{communication_language}:> Notice: bmad-create-prd is deprecated and will be removed in a future release. It now forwards to bmad-prd with create intent. To silence this notice and access the full new customization surface (prd_template, validation_checklist_template, validation_report_template, doc_standards, finalize_reviewers, external_sources, external_handoffs, prd_output_path, run_folder_pattern), migrate _bmad/custom/bmad-create-prd.toml to _bmad/custom/bmad-prd.toml and invoke bmad-prd directly next time. Customization fields that were in this version still remain in the new version and will be respected if present in _bmad/custom/bmad-prd.toml, but the new version also supports additional fields that you can take advantage of by migrating.
bmad-prd with the following context. Pass these as the activating context so bmad-prd honors them instead of resolving its own customization from scratch:create — skip bmad-prd's usual intent detection step.bmad-prd's own customize.toml for the four legacy fields. For everything else (prd_template, validation_checklist_template, validation_report_template, doc_standards, finalize_reviewers, prd_output_path, run_folder_pattern, external_sources, external_handoffs), use bmad-prd's own defaults and overrides as normal:activation_steps_prepend = the resolved value from step 1activation_steps_append = the resolved value from step 1persistent_facts = the resolved value from step 1on_complete = the resolved value from step 1bmad-prd 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 | 16,360 | 116,939 | +615% | 1 | 1 | 0% | 2,467 | 2,037 | -17% | 0 | 0 | — |
case-02 | fail→pass | 10,279 | 12,344 | +20% | 1 | 1 | 0% | 1,309 | 2,206 | +69% | 0 | 0 | — |
case-03 | fail→fail | 84,394 | 13,097 | -84% | 1 | 1 | 0% | 4,561 | 1,074 | -76% | 0 | 0 | — |
case-04 | fail→pass | 13,723 | 16,525 | +20% | 1 | 1 | 0% | 2,096 | 2,226 | +6% | 0 | 0 | — |
case-05 | fail→fail | 10,830 | 5,077 | -53% | 1 | 1 | 0% | 1,523 | 1,537 | +1% | 0 | 0 | — |
case-06 | pass→pass | 23,846 | 4,363 | -82% | 1 | 1 | 0% | 1,781 | 1,470 | -17% | 0 | 0 | — |
case-07 | fail→pass | 13,293 | 5,778 | -57% | 1 | 1 | 0% | 1,835 | 1,627 | -11% | 0 | 0 | — |
case-08 | pass→pass | 10,943 | 4,278 | -61% | 1 | 1 | 0% | 1,863 | 1,363 | -27% | 0 | 0 | — |
case-09 | fail→pass | 11,116 | 3,029 | -73% | 1 | 1 | 0% | 1,313 | 1,182 | -10% | 0 | 0 | — |
case-10 | fail→pass | 9,160 | 6,065 | -34% | 1 | 1 | 0% | 1,307 | 1,541 | +18% | 0 | 0 | — |
case-11 | fail→pass | 13,923 | 6,049 | -57% | 1 | 1 | 0% | 1,927 | 1,994 | +3% | 0 | 0 | — |
case-12 | fail→pass | 17,534 | 36,978 | +111% | 1 | 1 | 0% | 1,874 | 1,933 | +3% | 0 | 0 | — |
case-13 | fail→pass | 13,736 | 12,838 | -7% | 1 | 1 | 0% | 2,507 | 2,735 | +9% | 0 | 0 | — |
case-14 | fail→pass | 9,549 | 7,671 | -20% | 1 | 1 | 0% | 1,411 | 1,909 | +35% | 0 | 0 | — |
case-15 | fail→pass | 25,859 | 9,019 | -65% | 1 | 1 | 0% | 2,617 | 2,276 | -13% | 0 | 0 | — |
case-16 | pass→pass | 11,036 | 7,108 | -36% | 1 | 1 | 0% | 1,586 | 2,001 | +26% | 0 | 0 | — |
case-17 | fail→pass | 9,738 | 6,534 | -33% | 1 | 1 | 0% | 1,337 | 1,886 | +41% | 0 | 0 | — |
case-18 | fail→pass | 24,663 | 6,617 | -73% | 1 | 1 | 0% | 1,926 | 1,743 | -10% | 0 | 0 | — |
case-19 | fail→fail | 4,687 | 13,320 | +184% | 1 | 1 | 0% | 666 | 2,939 | +341% | 0 | 0 | — |
case-20 | fail→fail | 11,830 | 11,612 | -2% | 1 | 1 | 0% | 1,776 | 2,987 | +68% | 0 | 0 | — |
case-21 | pass→pass | 15,818 | 18,866 | +19% | 1 | 1 | 0% | 2,820 | 2,739 | -3% | 0 | 0 | — |
case-22 | pass→pass | 7,320 | 3,852 | -47% | 1 | 1 | 0% | 1,103 | 1,332 | +21% | 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 21 counted toward the lift figure. The other 1 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 +59 percentage points is the difference between those two pass rates over the 21 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.
Other measured skills in the registry, with their headline benchmark lift.