research-direction-scouting

name	research-direction-scouting
description	Scout and rank promising research directions in EECS using novelty, feasibility, available tooling, data access, benchmark fit, and publication risk. Use when choosing a thesis direction, starting a new project, framing a workshop or conference submission, or deciding whether an idea is incremental, premature, or genuinely worth pursuing.

Research Direction Scouting

Use this skill before a project gets expensive.

Core Workflow

1. Define the field, target venues, available compute or hardware, timeline, and collaboration constraints.
2. Build a short map of saturated, emerging, and underexplored subproblems.
3. Compare candidate ideas on:
   • novelty,
   • tractability,
   • resource fit,
   • evaluation clarity,
   • releaseability.
4. Separate flashy ideas from executable research plans.
5. End with one primary direction and one backup direction.

Execution Rules

1. Never score novelty without comparing to the last 2 to 3 years of papers.
2. Penalize ideas with vague evaluation or inaccessible hardware.
3. Prefer directions with clear falsification criteria.
4. Be explicit about whether an idea is a paper, a system, a benchmark, or a tool.

Output Contract

Return:

1. Direction table.
2. Ranking rationale.
3. Primary recommendation.
4. Fast validation experiment.
5. Main risks and fallback.