Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Domain-specific configuration profiles for learning resource creation. Defines search strategies, special fields, terminology policies, and content structures for different academic domains: technology, history, science, arts, and general. Use when researcher or writer agents need domain-adapted behavior.
.claude/skills/aiskillstore-domain-profiles/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | -25% | 0% |
| case-02 | ✗→✓ | ▲ Improved | -10% | 0% |
| case-03 | ✗→✓ | ▲ Improved | -22% | 0% |
| case-05 | ✗→✓ | ▲ Improved | -27% | 0% |
| case-06 | ✗→✓ | ▲ Improved | -15% | 0% |
This skill provides domain-specific configurations for learning resource creation.
/init Phase 2 (Research Collection)| Profile | File | Description | | ---------- | ------------------------------ | ---------------------------------------------- | | Technology | technology.md | Programming, frameworks, tools, APIs | | History | history.md | Historical events, periods, civilizations | | Science | science.md | Physics, chemistry, biology, mathematics | | Arts | arts.md | Visual arts, music, performing arts | | General | language.md | General topics, linguistics, language learning |
Note: The "general" domain uses language.md profile, which contains broadly applicable patterns for educational content.
Each profile contains:
Authoritative sources and search query patterns for the domain.
Domain-specific metadata fields to collect and include.
How to handle technical terms, translations, and citations.
Recommended document organization and pedagogical approach.
All agents should load domain profiles using this standardized pattern:
Read("skills/domain-profiles/{domain}.md")Domain to File Mapping:
| Input Domain | File to Read | |--------------|--------------| | technology | technology.md | | history | history.md | | science | science.md | | arts | arts.md | | general | language.md |
IMPORTANT: When domain is "general", agents MUST read language.md, not "general.md" (which doesn't exist).
| Agent | Sections to Extract | |-------|---------------------| | researcher | Search Strategy, Special Fields, Quality Indicators | | research-collector | Search Strategy, Special Fields | | writer | Content Structure, Terminology Policy | | reviewer | Review Criteria (Critical Checks, Quality Checks, Style Checks) |
# In researcher agent prompt
Read("skills/domain-profiles/technology.md")
# Extract Search Strategy section for domain-appropriate queries
# In writer agent prompt
Read("skills/domain-profiles/technology.md")
# Apply Content Structure and Terminology Policy to documentDomains are determined by the project-interviewer during the interview:
| Domain | Typical Topics | | ---------------- | -------------------------------------- | | technology | Python, React, Docker, API, 프로그래밍 | | history | 조선시대, 르네상스, 세계대전, 문명 | | science | 양자역학, 미적분, 세포생물학, 화학 | | arts | 유화, 작곡, 조각, 연기, 디자인 | | language/general | 언어학, 글쓰기, 일반 교양 |
If domain is unclear or "general", use:
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-04 | pass→pass | 9,884 | 7,702 | -22% | 1 | 1 | 0% | 1,658 | 1,238 | -25% | 0 | 0 | — |
case-01 | fail→pass | 12,479 | 9,098 | -27% | 1 | 1 | 0% | 1,990 | 1,496 | -25% | 0 | 0 | — |
case-02 | fail→pass | 14,510 | 8,119 | -44% | 1 | 1 | 0% | 1,402 | 1,266 | -10% | 0 | 0 | — |
case-03 | fail→pass | 11,243 | 3,765 | -67% | 1 | 1 | 0% | 1,828 | 1,428 | -22% | 0 | 0 | — |
case-05 | fail→pass | 15,924 | 8,362 | -47% | 1 | 1 | 0% | 1,722 | 1,259 | -27% | 0 | 0 | — |
case-06 | fail→pass | 8,063 | 7,586 | -6% | 1 | 1 | 0% | 1,359 | 1,159 | -15% | 0 | 0 | — |
case-07 | fail→pass | 7,740 | 2,377 | -69% | 1 | 1 | 0% | 1,136 | 1,133 | -0% | 0 | 0 | — |
case-08 | fail→pass | 18,466 | 10,169 | -45% | 1 | 1 | 0% | 2,688 | 1,540 | -43% | 0 | 0 | — |
case-09 | fail→pass | 18,255 | 2,378 | -87% | 1 | 1 | 0% | 1,927 | 1,153 | -40% | 0 | 0 | — |
case-10 | pass→pass | 11,320 | 6,943 | -39% | 1 | 1 | 0% | 945 | 1,063 | +12% | 0 | 0 | — |
case-11 | fail→pass | 6,740 | 2,516 | -63% | 1 | 1 | 0% | 1,174 | 1,187 | +1% | 0 | 0 | — |
case-12 | pass→pass | 8,577 | 2,707 | -68% | 1 | 1 | 0% | 1,316 | 1,224 | -7% | 0 | 0 | — |
case-13 | fail→pass | 7,675 | 7,589 | -1% | 1 | 1 | 0% | 1,193 | 1,158 | -3% | 0 | 0 | — |
case-14 | fail→pass | 16,348 | 7,283 | -55% | 1 | 1 | 0% | 1,803 | 1,146 | -36% | 0 | 0 | — |
case-15 | fail→pass | 12,172 | 7,585 | -38% | 1 | 1 | 0% | 1,270 | 1,192 | -6% | 0 | 0 | — |
case-16 | fail→pass | 6,602 | 8,098 | +23% | 1 | 1 | 0% | 1,114 | 1,290 | +16% | 0 | 0 | — |
case-17 | fail→pass | 12,631 | 7,809 | -38% | 1 | 1 | 0% | 1,169 | 1,195 | +2% | 0 | 0 | — |
case-18 | fail→pass | 12,766 | 7,340 | -43% | 1 | 1 | 0% | 1,964 | 1,151 | -41% | 0 | 0 | — |
case-19 | fail→pass | 4,999 | 3,571 | -29% | 1 | 1 | 0% | 830 | 1,331 | +60% | 0 | 0 | — |
case-20 | pass→fail | 13,892 | 7,922 | -43% | 1 | 1 | 0% | 1,475 | 2,092 | +42% | 0 | 0 | — |
case-21 | pass→pass | 12,796 | 14,618 | +14% | 1 | 1 | 0% | 2,071 | 3,079 | +49% | 0 | 0 | — |
case-22 | pass→pass | 14,115 | 10,284 | -27% | 1 | 1 | 0% | 1,629 | 2,563 | +57% | 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 +68 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.