nota-america/forgecat-agent-profiles / profiles/anthropics/knowledge-work-plugins/anthropics_knowledge-work-plugins_bio-research/for-forgecat/skills/scientific-problem-selection
scientific-problem-selection
VerifiedThis 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".
| File | Kind | Size |
|---|---|---|
| LICENSE.txt | markdown | 10.0 KB |
| SKILL.md | skill md | 10.9 KB |
| references/01-intuition-pumps.md | markdown | 12.2 KB |
| references/02-risk-assessment.md | markdown | 12.9 KB |
| references/03-optimization-function.md | markdown | 17.7 KB |
| references/04-parameter-strategy.md | markdown | 14.1 KB |
| references/05-decision-tree.md | markdown | 3.2 KB |
| references/06-adversity-planning.md | markdown | 4.6 KB |
| references/07-problem-inversion.md | markdown | 5.5 KB |
| references/08-integration-synthesis.md | markdown | 5.3 KB |
| references/09-meta-framework.md | markdown | 16.2 KB |
references/07-problem-inversion.md
View on GitHub# SKILL 7: Problem Inversion Strategies ("Turn It On Its Head")
## Overview
This skill provides three concrete strategies for navigating around obstacles by reframing problems. When stuck, instead of pushing harder on the current approach, try inverting the problem.
## Core Principle
**"Turn a problem on its head"**
Three powerful strategies:
1. **Unfix parameters** (covered in Skill 4, applied here in crisis)
2. **Don't achieve goal A? Achieve comparable goal B**
3. **"I have the answer; what is the question?"**
## Strategy 1: Unfix Parameters (In Crisis Mode)
**When to Use:** Run-of-the-mill issues in project execution
**Approach:** Let a "sacred" fixed parameter float
**Example from Lecture:**
- **Stuck:** Spatial transcriptomics of APC-T cell interactions in tumor microenvironment
- **All fixed:** Technique, cell types, context
- **Inversion:**
- Unfix technique → What else could measure these interactions?
- Unfix cell types → What other interactions matter in tumors?
- Unfix context → Where else do APC-T interactions matter?
**Your Application:**
For each fixed parameter in your project:
- What if this floated?
- What alternatives exist?
- Which would be easier/faster/more informative?
## Strategy 2: Comparable Goal Substitution
**When to Use:** Existential threats to project (can't achieve original goal)
**Approach:** Achieve a different but equally valuable goal
**Mindset Shift:**
- **Wrong:** "I failed to do X"
- **Right:** "The world needs Y instead, which I CAN do"
**Example from Lectures: PROTAC Story**
- **Goal A (Failed):** Degrade specific therapeutic target
- **Goal B (Achieved):** Map which kinases ARE degradable
- **Value:** B is more impactful (general principle + method validation)
- **Learning:** Target engagement ≠ degradation (important discovery)
**Framework:**
1. **Original goal:** [What you wanted]
2. **Why it failed:** [Specific reason]
3. **What CAN you do with current data/tools:** [Capabilities]
4. **Comparable goals:**
- Option 1: [Different but related goal]
- Option 2: [Another alternative]
- Option 3: [Yet another]
5. **Which is most valuable:** [Analysis]
6. **How to frame it:** [Communication strategy]
## Strategy 3: Answer Seeking Question
**When to Use:** End-of-project challenges (interpretation, framing, application)
**Approach:** You got an answer, but not to your original question. What question DOES your data answer?
**Mindset Shift:**
- **Wrong:** "This doesn't answer my question"
- **Right:** "What interesting question does this answer?"
**Example from Lectures: Steroid Receptor**
- **Original Question:** What is THE receptor for this steroid?
- **Answer Obtained:** Binds multiple receptors at different affinities
- **Problem:** Can't answer original question (no single receptor)
- **Inversion:** "What question does this answer?"
- **New Question:** How does finite receptor pool sense infinite lipids?
- **Answer:** Combinatorial sensing (pattern = unique "chord")
- **Impact:** More interesting than intended finding
**Framework:**
1. **Original question:** [What you asked]
2. **Data obtained:** [What you actually found]
3. **Why it doesn't answer:** [The mismatch]
4. **What DOES the data show clearly:** [Solid findings]
5. **What questions could these answer:**
- Question 1: [Option]
- Question 2: [Option]
- Question 3: [Option]
6. **Which is most interesting:** [Assessment]
7. **How to reframe paper/project:** [New framing]
## Workflow
### Phase 1: Identify Your Obstacle
- **Type:** Technical / Biological / Competitive / Interpretive
- **Severity:** Run-of-mill / Existential / End-stage
- **Description:** [What's blocking you]
### Phase 2: Select Strategy
| Obstacle Type | Recommended Strategy |
|--------------|---------------------|
| Technical barrier, mid-project | Strategy 1 (Unfix parameters) |
| Can't achieve original goal | Strategy 2 (Comparable goal) |
| Have data, unclear what it means | Strategy 3 (Answer seeking question) |
### Phase 3: Apply Strategy
Work through the relevant framework above with your specific situation.
### Phase 4: Evaluate Alternatives
For each alternative generated:
- **Scientific value:** How interesting is this?
- **Feasibility:** How hard to execute?
- **Timeline:** How long will it take?
- **Impact:** How does this compare to original plan?
- **Your advantage:** Do you still have edge here?
## Output: Problem Inversion Analysis
**Page 1: Current Situation**
- **Obstacle:** [Clear description]
- **Why you're stuck:** [Root cause]
- **Original plan:** [What you intended]
- **Current capability:** [What you CAN do]
**Page 2: Strategy Applications**
**Strategy 1 (Unfix Parameters):**
| Fixed Parameter | If This Floated | Alternative Approaches | Assessment |
|----------------|-----------------|----------------------|------------|
| [Param 1] | [Consequences] | [Options] | [Value] |
**Strategy 2 (Comparable Goals):**
| Original Goal | Why It Failed | Comparable Goal | Value Assessment |
|--------------|---------------|----------------|------------------|
| [Goal A] | [Reason] | [Goal B] | [Compare impact] |
**Strategy 3 (Answer → Question):**
- **Data obtained:** [What you have]
- **Question 1 it could answer:** [Option 1]
- **Question 2 it could answer:** [Option 2]
- **Question 3 it could answer:** [Option 3]
- **Most interesting:** [Selection + reasoning]
**Page 3: Recommended Path**
- **Selected strategy:** [1, 2, or 3]
- **New direction:** [Specific plan]
- **Why this is better:** [Not just "it works" but "it's more interesting"]
- **Communication approach:** [How to frame this pivot]
- **Timeline:** [New schedule]
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| 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: 472k | Updated | Claude CodeCodex | Verified | Install |
| 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: 322k | Updated | Claude CodeCodex | Verified | Install |
| dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependencies | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| 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 available | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |