agent analysis skill risk: low

ML Experiment Results Analyzer

The prompt asks the model to analyze ML experiment results by locating JSON/CSV files, building comparison tables with independent/dependent variables and deltas, performing statis…

SKILL 1 file

SKILL.md

Download

---
name: analyze-results
description: "Analyze ML experiment results, compute statistics, generate comparison tables and insights. Use when user says /\"analyze results/\", /\"compare/\", or needs to interpret experimental data."
---
# Analyze Experiment Results

Analyze: $ARGUMENTS

## Workflow

### Step 1: Locate Results
Find all relevant JSON/CSV result files:
- Check `figures/`, `results/`, or project-specific output directories
- Parse JSON results into structured data

### Step 2: Build Comparison Table
Organize results by:
- **Independent variables**: model type, hyperparameters, data config
- **Dependent variables**: primary metric (e.g., perplexity, accuracy, loss), secondary metrics
- **Delta vs baseline**: always compute relative improvement

### Step 3: Statistical Analysis
- If multiple seeds: report mean +/- std, check reproducibility
- If sweeping a parameter: identify trends (monotonic, U-shaped, plateau)
- Flag outliers or suspicious results

### Step 4: Generate Insights
For each finding, structure as:
1. **Observation**: what the data shows (with numbers)
2. **Interpretation**: why this might be happening
3. **Implication**: what this means for the research question
4. **Next step**: what experiment would test the interpretation

### Step 5: Update Documentation
If findings are significant:
- Propose updates to project notes or experiment reports
- Draft a concise finding statement (1-2 sentences)

## Output Format
Always include:
1. Raw data table
2. Key findings (numbered, concise)
3. Suggested next experiments (if any)

INPUTS

$ARGUMENTS REQUIRED: experiment results or references to analyze

REQUIRED CONTEXT

$ARGUMENTS
JSON/CSV result files in figures/ or results/

TOOLS REQUIRED

file_search

ROLES & RULES

always compute relative improvement
report mean +/- std, check reproducibility
identify trends (monotonic, U-shaped, plateau)
Flag outliers or suspicious results
Always include: 1. Raw data table 2. Key findings (numbered, concise) 3. Suggested next experiments (if any)

EXPECTED OUTPUT

Format

structured_report

Schema

markdown_sections · Raw data table, Key findings, Suggested next experiments

Constraints

always include raw data table
key findings numbered and concise
suggested next experiments if any

SUCCESS CRITERIA

Locate all relevant JSON/CSV result files
Build comparison table with independent/dependent variables and deltas
Perform statistical analysis
Generate insights with Observation, Interpretation, Implication, Next step
Update documentation if findings significant

FAILURE MODES

May miss result files outside figures/ or results/
May fail to detect reproducibility issues without multiple seeds

CAVEATS

Dependencies

$ARGUMENTS
figures/ or results/ directories

Missing context

Preferred output format details (e.g., markdown table style, file export).
How to handle cases with no result files found.

Ambiguities

Does not specify how $ARGUMENTS should be formatted or what it typically contains.
"project-specific output directories" is undefined and context-dependent.

QUALITY

OVERALL: 0.73
CLARITY: 0.82
SPECIFICITY: 0.68
REUSABILITY: 0.78
COMPLETENESS: 0.65

IMPROVEMENT SUGGESTIONS

Replace $ARGUMENTS with a clearer placeholder such as {{experiment_description}} and add usage examples.
Add explicit instruction on output medium (console, markdown file, etc.).

USAGE

Copy the prompt above and paste it into your AI of choice — Claude, ChatGPT, Gemini, or anywhere else you're working. Replace any placeholder sections with your own context, then ask for the output.