Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Transform PRDs (Product Requirements Documents) into structured XML app specifications optimized for AI coding agents. Converts developer-focused docs with code examples into declarative agent-consumable format. USE WHEN user says "convert PRD", "generate app spec", "transform PRD", "create specification from requirements", or wants to prepare a PRD for agent consumption.
.claude/skills/aiskillstore-prd-to-appspec/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-09 | ✗→✓ | ▲ Improved | -13% | 0% |
| case-11 | ✗→✓ | ▲ Improved | -19% | 0% |
| case-12 | ✗→✓ | ▲ Improved | -15% | 0% |
| case-13 | ✗→✓ | ▲ Improved | 22% | 0% |
| case-16 | ✗→✓ | ▲ Improved | 17% | 0% |
Transform Product Requirements Documents (PRDs) into structured XML application specifications optimized for AI coding agents.
Converts developer-focused PRDs (with code snippets, TDD plans, implementation details) into declarative XML specifications that AI coding agents can consume more effectively.
Input: PRD with technical details, code examples, architecture decisions Output: Structured app_spec.txt in XML format
Run the /convert-prd workflow, which provides:
| PRD Has | App Spec Gets | |---------|---------------| | Function implementations | Feature descriptions | | Pydantic field validators | Data constraints in prose | | Try/except patterns | Error handling requirements | | Test assertions | Success criteria | | CLI commands | API/command summaries | | Directory structure | Technology stack context |
xml<project_specification> <project_name>...</project_name> <overview>...</overview> <technology_stack>...</technology_stack> <core_features>...</core_features> <database_schema>...</database_schema> <api_endpoints_summary>...</api_endpoints_summary> <implementation_steps>...</implementation_steps> <success_criteria>...</success_criteria> </project_specification>
The app_spec tells an agent WHAT to build without dictating exact implementation.
For complete step-by-step instructions: workflows/convert-prd.md
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→fail | 12,416 | 2,732 | -78% | 1 | 1 | 0% | 2,272 | 839 | -63% | 0 | 0 | — |
case-02 | fail→fail | 18,814 | 7,549 | -60% | 1 | 1 | 0% | 3,810 | 1,784 | -53% | 0 | 0 | — |
case-03 | fail→fail | 40,153 | 3,298 | -92% | 1 | 1 | 0% | 1,110 | 724 | -35% | 0 | 0 | — |
case-04 | pass→pass | 25,358 | 13,946 | -45% | 1 | 1 | 0% | 4,038 | 2,798 | -31% | 0 | 0 | — |
case-05 | pass→fail | 26,560 | 6,315 | -76% | 1 | 1 | 0% | 6,179 | 1,575 | -75% | 0 | 0 | — |
case-06 | pass→pass | 11,350 | 26,833 | +136% | 1 | 1 | 0% | 2,058 | 5,652 | +175% | 0 | 0 | — |
case-07 | pass→pass | 14,197 | 7,026 | -51% | 1 | 1 | 0% | 2,311 | 1,764 | -24% | 0 | 0 | — |
case-08 | pass→pass | 12,805 | 6,241 | -51% | 1 | 1 | 0% | 2,002 | 1,550 | -23% | 0 | 0 | — |
case-09 | fail→pass | 12,316 | 7,838 | -36% | 1 | 1 | 0% | 2,037 | 1,780 | -13% | 0 | 0 | — |
case-10 | pass→pass | 13,195 | 6,662 | -50% | 1 | 1 | 0% | 2,085 | 1,636 | -22% | 0 | 0 | — |
case-11 | fail→pass | 12,129 | 7,287 | -40% | 1 | 1 | 0% | 2,189 | 1,778 | -19% | 0 | 0 | — |
case-12 | fail→pass | 12,427 | 6,731 | -46% | 1 | 1 | 0% | 1,947 | 1,655 | -15% | 0 | 0 | — |
case-13 | fail→pass | 4,941 | 2,981 | -40% | 1 | 1 | 0% | 793 | 965 | +22% | 0 | 0 | — |
case-14 | pass→pass | 8,405 | 5,051 | -40% | 1 | 1 | 0% | 1,313 | 1,267 | -4% | 0 | 0 | — |
case-15 | pass→fail | 5,411 | 2,596 | -52% | 1 | 1 | 0% | 808 | 917 | +13% | 0 | 0 | — |
case-16 | fail→pass | 4,375 | 1,411 | -68% | 1 | 1 | 0% | 628 | 732 | +17% | 0 | 0 | — |
case-17 | fail→pass | 15,160 | 1,935 | -87% | 1 | 1 | 0% | 2,159 | 765 | -65% | 0 | 0 | — |
case-18 | fail→pass | 11,009 | 3,954 | -64% | 1 | 1 | 0% | 1,764 | 1,147 | -35% | 0 | 0 | — |
case-19 | fail→pass | 6,195 | 2,705 | -56% | 1 | 1 | 0% | 1,023 | 960 | -6% | 0 | 0 | — |
case-20 | fail→pass | 11,469 | 2,749 | -76% | 1 | 1 | 0% | 1,775 | 994 | -44% | 0 | 0 | — |
case-21 | fail→fail | 9,765 | 2,055 | -79% | 1 | 1 | 0% | 1,569 | 861 | -45% | 0 | 0 | — |
case-22 | fail→pass | 8,070 | 1,901 | -76% | 1 | 1 | 0% | 1,287 | 766 | -40% | 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 +36 percentage points is the difference between those two pass rates over the 21 comparable cases. 4 cases got worse with the skill loaded, and they are 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.