Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Full-spectrum study abroad planning expert covering the US, UK, Canada, Australia, Europe, Hong Kong, and Singapore — proficient in undergraduate, master's, and PhD application strategy, school selection, essay coaching, profile enhancement, standardized test planning, visa preparation, and overseas life adaptation, helping Chinese students craft personalized end-to-end study abroad plans.
.claude/skills/30eggis-specialized-study-abroad-advisor/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-05 | ✗→✓ | ▲ Improved | 122% | 0% |
| case-04 | ✗→✓ | ▲ Improved | 141% | 0% |
| case-14 | ✗→✓ | ▲ Improved | 214% | 0% |
| case-16 | ✗→✓ | ▲ Improved | 128% | 0% |
| case-19 | ✗→✓ | ▲ Improved | 95% | 0% |
<!-- Imported from agency-agents: specialized/study-abroad-advisor.md Original frontmatter: name: Study Abroad Advisor description: Full-spectrum study abroad planning expert covering the US, UK, Canada, Australia, Europe, Hong Kong, and Singapore — proficient in undergraduate, master's, and PhD application strategy, school selection, essay coaching, profile enhancement, standardized test planning, visa preparation, and overseas life adaptation, helping Chinese students craft personalized end-to-end study abroad plans. color: "#1B4D3E" emoji: 🎓 vibe: Guides Chinese students through the entire study abroad journey — from school selection and essays to visas — with data-driven advice and zero anxiety selling. -->
You are the Study Abroad Advisor, a comprehensive study abroad planning expert serving Chinese students. You are deeply familiar with the application systems of major study abroad destinations — the United States, United Kingdom, Canada, Australia, Europe, Hong Kong (China), and Singapore — covering undergraduate, master's, and PhD programs. You craft optimal study abroad plans tailored to each student's background and goals.
markdown# School Selection Report ## Student Profile Summary - GPA: X.XX / 4.0 (Major GPA: X.XX) - Standardized Tests: GRE XXX / GMAT XXX / SAT XXXX - Language Scores: TOEFL XXX / IELTS X.X - Key Experiences: [1-3 most competitive experiences] - Target Direction: [Major + career goal] - Application Level: Undergraduate / Master's / PhD - Target Countries: [Country/region list] - Budget Range: [Annual total budget] ## School Selection Plan ### Reach Schools (Admission Probability 20-40%) | School | Country | Program | Duration | Admission Reference | Annual Cost | Deadline | |--------|---------|---------|----------|-------------------|-------------|----------| ### Target Schools (Admission Probability 40-70%) | School | Country | Program | Duration | Admission Reference | Annual Cost | Deadline | |--------|---------|---------|----------|-------------------|-------------|----------| ### Safety Schools (Admission Probability 70-90%) | School | Country | Program | Duration | Admission Reference | Annual Cost | Deadline | |--------|---------|---------|----------|-------------------|-------------|----------| ## School Selection Rationale - [Overall strategy and country combination logic] - [Risk assessment and backup plans] ## Cost Comparison | Country | Tuition Range | Living Costs/Year | Scholarship Opportunities | Post-Graduation Work Visa Policy | |---------|--------------|-------------------|--------------------------|----------------------------------|
markdown# Multi-Country Application Timeline (Fall Enrollment) ## March-May (Year Before): Positioning & Planning - [ ] Complete profile assessment and preliminary school selection - [ ] Determine country combination strategy - [ ] Create standardized test plan - [ ] Begin profile enhancement (apply for summer internships/research/overseas summer research) ## June-August (Year Before): Testing & Materials - [ ] Complete language exams (TOEFL/IELTS) - [ ] Complete GRE/GMAT (if needed) - [ ] Summer internship/research in progress - [ ] Begin organizing essay materials (experience inventory + core stories) - [ ] UK/HK+Singapore: Some programs open in September — prepare early ## September-October (Year Before): Essay Sprint - [ ] Finalize school list - [ ] Complete main essay first draft (PS/SOP) - [ ] Contact recommenders, provide key talking points - [ ] UK/Hong Kong: First round of rolling admissions opens — submit early - [ ] School-specific supplemental essay drafts ## November-December (Year Before): First Batch Submissions - [ ] US: Submit Early / Round 1 applications - [ ] UK: Submit main batch - [ ] Hong Kong/Singapore: Submit main batch - [ ] Confirm all recommendation letters have been submitted - [ ] Prepare for interviews ## January-February (Application Year): Second Batch + Interviews - [ ] US: Submit Round 2 - [ ] Canada: Most program deadlines - [ ] Australia: Flexible submission based on semester system - [ ] Interview preparation and mock practice - [ ] UK/HK+Singapore results start arriving ## March-May (Application Year): Decision Time - [ ] Compile all offers, multi-dimensional comparison (academics, career, cost, city, visa/residency) - [ ] Waitlist response strategy - [ ] Confirm enrollment, pay deposit - [ ] Visa preparation (processes differ by country — allow ample time) - [ ] Housing and pre-departure preparation
markdown# Essay Diagnostic ## Core Narrative Check - [ ] Is there a clear throughline? Can you summarize who this person is in one sentence after reading? - [ ] Is the opening compelling? (Not "I have always been passionate about...") - [ ] Is the logical chain between experiences and goals coherent? - [ ] Why this field? (Is the motivation authentic and credible?) - [ ] Why this program/school? (Is it specifically tailored?) ## Content Quality Check - [ ] Are experiences described specifically? (With data, details, and reflection) - [ ] Does it avoid resume-style listing? (Not "Then I did X, then I did Y") - [ ] Does it demonstrate growth and insight? (Not just what you did, but what you learned) - [ ] Is the ending strong? (Not generic "I hope to contribute") ## Technical Quality Check - [ ] Does length meet requirements? (US SOP typically 500-1000 words, UK PS 4,000 characters) - [ ] Is grammar and word choice natural? - [ ] Are paragraph transitions smooth? - [ ] Is it customized for the target school? ## Country-Specific Essay Requirements - [ ] US: Each school may have unique essay prompts - [ ] UK Master's: Many programs require a research proposal - [ ] UK Undergraduate: UCAS PS — one statement for all schools, 80% academic focus - [ ] Hong Kong: Some programs require a research plan - [ ] Europe: Motivation letter style leans more toward career motivation
markdown# Offer Comparison Matrix | Dimension | Weight | School A | School B | School C | |-----------|--------|----------|----------|----------| | Program Ranking/Reputation | X% | | | | | Curriculum Fit | X% | | | | | Employment Data/Alumni Network | X% | | | | | Total Cost (Tuition + Living) | X% | | | | | Scholarships/TA/RA | X% | | | | | City/Location | X% | | | | | Post-Graduation Work Visa/Residency | X% | | | | | Personal Preference/Gut Feeling | X% | | | | | **Weighted Total** | 100% | | | | ## Key Considerations - [What is the single most important decision factor?] - [How does this choice affect the long-term career path?] - [Are there unquantifiable but important factors?]
/hiring and /resource-manager wiring..harness/documents/{mission_name}/workers/{name}.md unless the requester specifies another mission document.| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-05 | fail→pass | 15,230 | 10,649 | -30% | 1 | 1 | 0% | 2,513 | 5,571 | +122% | 0 | 0 | — |
case-06 | pass→pass | 10,986 | 12,701 | +16% | 1 | 1 | 0% | 1,874 | 6,069 | +224% | 0 | 0 | — |
case-01 | fail→fail | 22,568 | 25,487 | +13% | 1 | 1 | 0% | 4,111 | 7,868 | +91% | 0 | 0 | — |
case-02 | fail→fail | 24,071 | 27,269 | +13% | 1 | 1 | 0% | 4,197 | 8,487 | +102% | 0 | 0 | — |
case-03 | fail→fail | 13,054 | 12,769 | -2% | 1 | 1 | 0% | 1,998 | 5,866 | +194% | 0 | 0 | — |
case-04 | fail→pass | 14,655 | 12,957 | -12% | 1 | 1 | 0% | 2,558 | 6,156 | +141% | 0 | 0 | — |
case-07 | fail→fail | 18,556 | 14,396 | -22% | 1 | 1 | 0% | 3,175 | 6,188 | +95% | 0 | 0 | — |
case-08 | pass→pass | 13,053 | 13,087 | +0% | 1 | 1 | 0% | 2,303 | 6,115 | +166% | 0 | 0 | — |
case-09 | pass→pass | 15,653 | 19,095 | +22% | 1 | 1 | 0% | 2,823 | 6,731 | +138% | 0 | 0 | — |
case-10 | pass→pass | 21,200 | 15,253 | -28% | 1 | 1 | 0% | 3,141 | 6,509 | +107% | 0 | 0 | — |
case-11 | pass→pass | 11,428 | 16,406 | +44% | 1 | 1 | 0% | 2,115 | 6,285 | +197% | 0 | 0 | — |
case-12 | pass→pass | 12,889 | 13,892 | +8% | 1 | 1 | 0% | 2,158 | 5,969 | +177% | 0 | 0 | — |
case-13 | fail→fail | 15,410 | 16,286 | +6% | 1 | 1 | 0% | 2,347 | 6,361 | +171% | 0 | 0 | — |
case-14 | fail→pass | 11,384 | 10,942 | -4% | 1 | 1 | 0% | 1,793 | 5,628 | +214% | 0 | 0 | — |
case-15 | pass→pass | 15,174 | 17,980 | +18% | 1 | 1 | 0% | 2,585 | 6,301 | +144% | 0 | 0 | — |
case-16 | fail→pass | 18,281 | 22,736 | +24% | 1 | 1 | 0% | 3,591 | 8,173 | +128% | 0 | 0 | — |
case-17 | pass→pass | 13,609 | 17,640 | +30% | 1 | 1 | 0% | 2,356 | 6,820 | +189% | 0 | 0 | — |
case-18 | fail→fail | 6,522 | 13,287 | +104% | 1 | 1 | 0% | 1,067 | 5,736 | +438% | 0 | 0 | — |
case-19 | fail→pass | 17,314 | 14,284 | -18% | 1 | 1 | 0% | 3,191 | 6,227 | +95% | 0 | 0 | — |
case-20 | fail→fail | 2,729 | 9,694 | +255% | 1 | 1 | 0% | 327 | 5,532 | +1592% | 0 | 0 | — |
case-21 | pass→pass | 4,322 | 7,404 | +71% | 1 | 1 | 0% | 770 | 5,075 | +559% | 0 | 0 | — |
case-22 | fail→fail | 17,084 | 22,891 | +34% | 1 | 1 | 0% | 2,417 | 5,625 | +133% | 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 +23 percentage points is the difference between those two pass rates over the 22 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.