sciagent-skill-creator
Scaffold a new SciAgent-Skills entry. Picks pipeline/toolkit/database/guide template, creates skills/{category}/{name}/SKILL.md with valid frontmatter, appends…
Structured ideation methods: SCAMPER, Six Thinking Hats, Morphological Analysis, TRIZ, Biomimicry, plus more. Decision framework for picking methods by challenge type (stuck, improving, systematic exploration, contradiction). Use when generating research ideas or exploring
$ npx -y skills add jaechang-hits/SciAgent-Skills --skill scientific-brainstorming --agent claude-codeHow it fires
How this skill gets triggered: by you, by Claude, or both.
/scientific-brainstormingContext preview
The summary Claude sees to decide when to auto-load this skill.
Structured ideation methods: SCAMPER, Six Thinking Hats, Morphological Analysis, TRIZ, Biomimicry, plus more. Decision framework for picking methods by challenge type (stuck, improving, systematic exploration, contradiction). Use when generating research ideas or exploring
name: "scientific-brainstorming" description: "Structured ideation methods: SCAMPER, Six Thinking Hats, Morphological Analysis, TRIZ, Biomimicry, plus more. Decision framework for picking methods by challenge type (stuck, improving, systematic exploration, contradiction). Use when generating research ideas or exploring interdisciplinary connections." license: "CC-BY-4.0"
Scientific brainstorming is a structured ideation process for generating, connecting, and evaluating research ideas. Unlike casual brainstorming, scientific brainstorming applies formal methodologies (SCAMPER, TRIZ, Morphological Analysis, etc.) matched to the specific creative challenge. The process moves through divergent exploration, connection-making, critical evaluation, and synthesis to produce actionable research directions with testable hypotheses.
Five principles guide effective scientific brainstorming sessions:
1. **Collaborative**: Brainstorming works best as dialogue, not monologue. Build on each other's ideas rather than presenting finished thoughts. Use "Yes, and..." framing to extend ideas before evaluating them. In AI-assisted sessions, the scientist contributes domain expertise and the AI contributes breadth and pattern-matching across disciplines.
2. **Curious**: Approach the problem space with genuine curiosity. Ask "what if" and "why not" before "why." Suspend expertise-driven assumptions temporarily to allow unexpected connections. Experts often dismiss novel directions because they conflict with established mental models -- curiosity counteracts this.
3. **Domain-Aware**: Ground brainstorming in real scientific constraints. Ideas must eventually connect to testable hypotheses, available methods, and feasible experiments. Domain knowledge channels creativity productively. Pure creativity without domain grounding produces ideas that cannot be tested; pure domain expertise without creativity produces incremental work.
4. **Structured**: Use formal ideation methods rather than unguided free association. Structure prevents cognitive fixation (repeatedly returning to the same idea space) and ensures systematic coverage of the possibility space. Unstructured brainstorming sessions typically explore less than 20% of the available idea space.
5. **Challenging**: Actively seek ideas that feel uncomfortable or counterintuitive. The most productive brainstorming sessions push past obvious solutions into territory that requires deeper analysis. If every idea generated feels reasonable and safe, the session is not pushing hard enough.
Nine structured methodologies, each suited to different creative challenges:
| Method | When to Use | Key Technique | Scientific Example | |--------|-------------|---------------|-------------------| | **SCAMPER** | Improving or extending an existing method/system | Systematically apply 7 operators (Substitute, Combine, Adapt, Modify, Put to use, Eliminate, Reverse) to the current approach | Substitute fluorescence for radioactive labeling in an assay; Combine two biomarker panels into a multiplex panel | | **Six Thinking Hats** | Need multiple perspectives on a research question | Assign structured roles: White (data/facts), Red (intuition/feelings), Black (critical/risks), Yellow (benefits/optimism), Green (creative alternatives), Blue (process management) | Evaluate a proposed clinical trial: White examines prior data, Black identifies ethical risks, Green suggests novel endpoints | | **Morphological Analysis** | Exploring all combinations within a design space | Define dimensions of the problem, list options per dimension, systematically explore combinations | Drug delivery: dimensions = carrier (liposome, nanoparticle, hydrogel) x targeting (passive, active, magnetic) x release (pH, thermal, enzymatic) | | **TRIZ** | Resolving technical contradictions | Identify the contradiction (improving X worsens Y), apply inventive principles, envision the ideal final result | Increasing drug potency (desired) increases toxicity (undesired) -- apply separation principle: target only affected tissue | | **Biomimicry** | Seeking nature-inspired solutions | Define function, biologize the question, discover natural models, abstract the principle, apply to problem | "How does nature filter particles?" leads to studying kidney nephrons for microfluidic filter design | | **Provocation (Po)** | Breaking out of fixed thinking patterns | State an impossible or absurd premise ("Po: cells never divide"), then extract useful principles from the provocation | "Po: proteins fold instantly" -- what if we engineered ultrafast folding domains? Leads to intrinsically disordered protein research | | **Random Input** | Need fresh connections when stuck in a rut | Select a random stimulus (word, image, object from nature), force connections to the research problem | Random word "bridge" + enzyme kinetics = bridging molecules that connect substrate to enzyme active site | | **Reverse Assumptions** | Questioning fundamental assumptions | List all assumptions about the problem, flip each one, explore consequences of each reversal | Assumption: "higher purity improves results" -- reverse: what if impurities are functional? Leads to studying beneficial contaminants | | **Future Backwards** | Envisioning long-term research directions | Start from a solved future state, work backwards to identify necessary intermediate steps and breakthroughs | "Cancer is cured in 2050" -- what needed to happen in 2040? 2030? What research today enables the 2030 milestone? |
The seven SCAMPER operators applied to scientific contexts:
What material, reagent, model organism, or technique could replace the current one? Example: replace mouse models with organoids; substitute CRISPR for siRNA knockdown.
Turn your AI coding agent into a life sciences expert — 199 bioinformatics skills for Claude Code covering RNA-seq, single-cell analysis, genomics, proteomics, drug discovery, and more. Boosted BixBench from 65% to 92%. Open source.
Scaffold a new SciAgent-Skills entry. Picks pipeline/toolkit/database/guide template, creates skills/{category}/{name}/SKILL.md with valid frontmatter, appends…
Bayesian modeling with PyMC 5: priors, likelihood, NUTS/ADVI sampling, diagnostics (R-hat, ESS), LOO/WAIC comparison, prediction. Hierarchical, logistic, GP…
Time-to-event modeling with scikit-survival: Cox PH (elastic net), Random Survival Forests, Boosting, SVMs for censored data. C-index, Brier, time-dependent…
Guided statistical analysis: test choice, assumption checks, effect sizes, power, APA reporting. Pick tests, verify assumptions, or format results for…
Python statistical modeling: regression (OLS, WLS, GLM), discrete (Logit, Poisson, NegBin), time series (ARIMA, SARIMAX, VAR), with rigorous inference,…
DL cell/nucleus segmentation for fluorescence and brightfield microscopy. Pre-trained models (cyto3, nuclei, tissuenet) and a generalist flow-based algorithm…