Skip to content

nota-america/forgecat-agent-profiles / profiles/anthropics/knowledge-work-plugins/anthropics_knowledge-work-plugins_bio-research/for-claude/.claude/skills/scientific-problem-selection

scientific-problem-selection

Verified

This skill should be used when scientists need help with research problem selection, project ideation, troubleshooting stuck projects, or strategic scientific decisions. Use this skill when users ask to pitch a new research idea, work through a project problem, evaluate project risks, plan research strategy, navigate decision trees, or get help choosing what scientific problem to work on. Typical requests include "I have an idea for a project", "I'm stuck on my research", "help me evaluate this project", "what should I work on", or "I need strategic advice about my research".

references/05-decision-tree.md

View on GitHub
# SKILL 5: Decision Tree Navigation ("The Altitude Dance")

## Overview
This skill teaches you to move fluidly between execution (Level 1: getting stuff done) and strategic evaluation (Level 2: critical thinking). Projects rarely unfold linearly—they require frequent course correction. Most trainees should spend MORE time on their project's decision tree.

## Core Principle
**"Learn the altitude dance"**

Move back and forth frequently between:
- **Level 1:** Full immersion in experimental details or coding
- **Level 2:** Step back, clear your head, evaluate as if someone else did the work

These cannot be done simultaneously. The key to navigating a project's decision tree is alternating between these levels deliberately.

## Key Concepts

**Why Decision Trees Matter:**
Once you're in a project, the landscape changes:
- You've learned from initial experiments
- New papers have been published
- Technology has advanced
- Your assumptions have been tested

At any decision point, you should rarely follow your plan from 2 years ago—there will be a better alternative.

**The Altitude Levels:**
- **Level 1 (Ground Level):** Doing the work, troubleshooting, optimizing
- **Level 2 (Strategic Altitude):** What did we learn? What should we do next?
- **Level 3 (Field Altitude):** How does this fit in the broader landscape?
- **Level 4 (Career Altitude):** Is this the right use of my finite time?

**Common Failure Modes:**
1. **Stuck in Level 1:** Troubleshooting endlessly without reassessing the plan
2. **Only Level 2:** Brilliant strategist but never rolls up sleeves
3. **No rhythm:** Switching randomly instead of deliberately

## Workflow

### Phase 1: Map Your Decision Tree

For your project, identify:
1. **Initial plan:** What was the intended path?
2. **Branch points:** Where might alternative paths emerge?
3. **Decision criteria:** What determines which branch to take?
4. **New information:** What could change the landscape?

### Phase 2: Establish Your Rhythm

**Recommended Schedule:**
- **Daily:** Level 1 work (experiments, coding, analysis)
- **Weekly:** Level 2 evaluation (1-2 hours, ideally Friday afternoon)
- **Monthly:** Level 3 field review (read new papers, attend seminars)
- **Quarterly:** Level 4 career check-in (with mentor)

**Level 2 Weekly Protocol:**
1. Clear your head (walk, coffee, change of scene)
2. Review what happened this week
3. Ask: What did we learn?
4. Ask: What should happen next?
5. Update decision tree
6. Plan next week's Level 1 work

### Phase 3: Decision Points

At each major branch point:

**Example: Genetic Screen Hits Wall**

Instead of endless troubleshooting:
- **Alternative 1:** Redo computational analysis with larger genome dataset
- **Alternative 2:** Use AlphaFold models to search for similar folds
- **Alternative 3:** Print and test larger candidate set (DNA synthesis cheaper now)

**Framework:**
1. **Acknowledge the stuck point**
2. **Step to Level 2:** Evaluate with fresh eyes
3. **Consider: What's newly possible?** (technology, knowledge)
4. **Generate 3 alternatives**
5. **Decide:** Troubleshoot more vs. pursue alternative

## Output: Decision Tree Map
- Visual map of your project's decision points
- Update frequency schedule
- Criteria for each branch point
- Protocol for getting unstuck
reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis.Stars: 472kUpdated Claude CodeCodexVerifiedInstall
diagnosing-superpowersobra/superpowersUse when a superpowers session went wrong and your human partner wants to know why — repeated work, ignored plans, stumbles, poor results, a skill that didn't fire, "it took too long", "why is it so expensive", "what is it doing" — or wants to build a bug report for the superpowers maintainers, for the current session or a past one identified by id or path, on any harness.Stars: 322kUpdated Claude CodeCodexVerifiedInstall
dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependenciesStars: 322kUpdated Claude CodeCodexVerifiedInstall
executing-plansobra/superpowersUse when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is availableStars: 322kUpdated Claude CodeCodexVerifiedInstall