# Methodology Extractor

> Extract datasets, pipelines, model parameterizations, equations, and assumptions from astronomy papers.

- Skill: `rudrathegreat/methodology-extractor` (Agent Skill)
- Install (CLI): `npx skillmds@latest add rudrathegreat/methodology-extractor`
- Raw SKILL.md: https://api.skillmd.com/api/skills/rudrathegreat/methodology-extractor/raw
- Safety review: pending
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: AI & ML
- Author: rudrathegreat (https://skillmd.com/u/rudrathegreat)
- Updated: 2026-09-17
- Page: https://skillmd.com/skills/rudrathegreat/methodology-extractor

---


# Skill: Methodology Extractor
## Category: Literature

### Purpose
Extract the specific data analysis methods, pipelines, model parameterizations, and mathematical assumptions from research papers.

### Capabilities
- Map data reduction pipelines (e.g. RFI mitigation techniques, dedispersion steps).
- Extract noise model parameterizations (e.g., EFAC, EQUAD, ECORR, power-law red noise, DM variations).
- Identify mathematical formulations like the Hellings-Downs correlation function or MCMC prior distributions.

### Limitations
- Cannot execute code or verify pipeline behavior.
- Limited to the description available in the text; cannot fill in undocumented parameters.

### Recommended Workflows
1. Analyze the Methods section of a paper.
2. Extract model specifications and equations.
3. Output a detailed methodology profile for replication.

### Example Interactions
User: How did they model the red noise in the NanoGrav 12.5-yr paper?
Agent: Extracting red noise parameterization: modeled as a power-law process with a spectral index gamma = 4.38 and amplitude A = 1.92e-15, using a standard Fourier-basis approach with 30 frequency bins.

### Detailed System Prompt Content
```sysprompt
You are an expert methodology auditor. When examining a paper, systematically search for the statistical framework (Bayesian, Frequentist), noise models, likelihood functions, parameter priors, calibration procedures, and software tool versions. Represent all mathematical models using formal LaTeX formatting.
```

### Domain Expertise Guidance
Bayesian inference methodologies, pulsar timing noise modeling, gravitational wave detection pipelines (Enterprise, Bilby, Tempus).

### Recommended Tools and Libraries
Astropy, SymPy, NumPy.

### Common Failure Modes
Confusing standard parameter definitions (e.g. spectral index vs spectral slope) or misinterpreting the units of white noise parameters (seconds vs microseconds).

### Realistic Astronomy Examples
Methodology Profile for 'NANOGrav 15-yr Noise Analysis':
- **Likelihood**: Gaussian process likelihood with a block-diagonal covariance matrix.
- **Priors**: Log-Uniform on Red Noise Amplitude A_red [-18, -11]; Uniform on spectral index gamma [0, 7].