Skip to content
OpenAgentSkill
Browsing is open · Catalog captured 2026-11-21 · Automatic updates and account services are paused.
Search & research

us-program-research

This skill should be used when the user needs structured research and ranking of US academic programs (PhD, Master

ericgandrade · Static checks on file

Installation

Read the source and dependencies before running a command. Copying does not install or execute anything.

npx skills add ericgandrade/claude-superskills --skill us-program-research

About this skill


name: us-program-research description: This skill should be used when the user needs structured research and ranking of US academic programs (PhD, Master license: MIT

US Academic Program Research — Complete Workflow

Purpose

Run structured research on US academic programs with credential analysis, parallel discovery, adaptive scorecards, and generation of an actionable application plan in a deliverable format.

When to Use

Use this skill when the task requires:

  • Selecting and ranking programs (PhD, Master's, or Bachelor's) in the US
  • Detailed comparison of curriculum, costs, and admission requirements
  • Application strategy based on profile, budget, and timeline
  • A final consolidated document for decision-making and execution

Progress Tracking

Display progress before each research phase:

[███░░░░░░░░░░░░░░░░░] 15% — Phase 0: Program Type Detection & Input Collection
[██████░░░░░░░░░░░░░░] 30% — Phase 1: Profile & Credential Analysis
[█████████░░░░░░░░░░░] 45% — Phase 2: Parallel Discovery (4 subagents)
[████████████░░░░░░░░] 60% — Phase 3: Parallel Deep Research (4 subagents)
[███████████████░░░░░] 75% — Phase 4: Adaptive Scorecards & Ranking
[█████████████████░░░] 85% — Phase 5: Document Generation
[████████████████████] 100% — Phase 6: Inline Report & Recommendations

Workflow

Follow the phases defined below in sequence, maintaining source traceability and separating facts from inferences.

Personalized research on US academic programs (PhD, Master's MS/MBA/MPS, or Bachelor's). Analyzes the candidate's profile, runs parallel searches via subagents, identifies hidden gems, applies an adaptive scorecard, and generates a complete ACTION_PLAN.md with rankings, curricula, costs, and a step-by-step checklist.

Output language: Portuguese (matches user research context). Queries/subagents: English (required for search quality).


Execution Instructions

  1. PHASE 0 — detect program type (FIRST QUESTION)
  2. PHASE 0B — collect all inputs before researching
  3. PHASE 2 — launch 4 subagents in ONE message (true parallelism)
  4. PHASE 3 — launch 4 deep-research subagents in ONE message
  5. PHASE 4 — apply adaptive scorecard after all research is complete
  6. PHASE 5 — generate full document and save
  7. PHASE 6 — present inline report to user

CRITICAL: All search queries and subagent prompts MUST be in English.


PHASE 0 — Program Type Detection

Use AskUserQuestion with: "What type of US academic program are you looking for?"

Options:

  1. PhD / Doctorate — 4–7 years of research. Usually fully-funded (stipend + tuition).
  2. Master's — MS, MBA, MPS, MEng (1.5–3 years). Executive, technical, online or in-person.
  3. Bachelor's / Undergraduate — First or second degree (BS, BA). 2–4 years.

The entire workflow adapts to this answer.


PHASE 0B — Input Collection (2 rounds of AskUserQuestion)

Round A — Personal profile (all types, max 4 questions):

  1. Full name + email
  2. Country of origin + language(s) of diplomas
  3. Highest current degree: institution, diploma type, GPA/grade (e.g., "8.1/10")
  4. Desired field of study (e.g., Computer Science, Business Analytics, AI, Finance...)

Round B — Logistics and preferences (max 4 questions):

  1. Target city/state in the US (or "open to any location")
  2. Preferred format: in-person / 100% online / hybrid / no preference
  3. Total budget: under $20k / $20–50k / $50–80k / over $80k / seeking funding
  4. Desired start date: 2026 / 2027 / 2028 / flexible

Additional type-specific question:

  • PhD: Specific research area? Known advisors? Funded programs only?
  • Master's: Current role/level (Junior/Manager/Director/VP/C-Level)? Will you work during the program?
  • Bachelor's: First or second degree? Transferable credits? Large/small campus preference?

Optional inputs (offer, do not require): CV, academic transcripts, prior executive courses, US immigration status (Green Card / F1-OPT / H1-B / other).


PHASE 1 — Profile & Credential Analysis

Determine foreign degree equivalency and evaluation strategy BEFORE scoring any program.

Action: Consult references/credential-analysis.md for:

  • Degree equivalency table by country (includes risk for Brazilian Tecnólogo degree)
  • WES vs ECE comparison by program type
  • WES ICAP step-by-step logistics (8 steps + 2 scenarios: favorable vs unfavorable result)
  • Anti-downgrading analysis for VP/C-Level candidates (Master's only)
  • Advisor fit analysis for PhD + contact email template
  • In-state tuition logic (Green Card + 12-month domicile = eligible)

Brazilian Tecnólogo = HIGH RISK. WES may evaluate as Associate's instead of Bachelor's. Recommend institutional letter + ECE as second opinion if WES result is unfavorable.


PHASE 2 — Parallel Discovery (4 Subagents)

CRITICAL: Launch all 4 subagents in ONE SINGLE message (Task tool × 4 simultaneous).

Divide research among 4 subagents with subagent_type="general-purpose":

SubagentFocusMinimum Programs
ARegional / Local In-Person4
BNational Online / Brand Equity5
CHidden Gems (low cost)3
DReference RankingsTop 20 context

Action: Consult references/subagent-prompts.md for the full prompts (in English) for each subagent, including all fields to collect and search queries.

After all 4 subagents complete:

  • Deduplicate programs (same school may appear in multiple subagents)
  • Verify accreditation: AACSB (business), ABET (engineering) — absence penalizes Brand Equity
  • For PhD: verify if fully-funded vs self-funded
  • Classify into groups: 1 (primary in-person), 1.5 (online brand equity), 2 (technical/deprioritized)
  • Target: 12–20 unique programs for Phase 3

PHASE 3 — Parallel Deep Research (4 Subagents)

**CRITICAL

Read the complete source ↗