# Comparison

> Compare entities or sources across dimensions. Use when the user wants a side-by-side assessment, not two separate facts.

- Skill: `samraddhashrivastavatech/comparison` (Agent Skill)
- Install (CLI): `npx skillmds@latest add samraddhashrivastavatech/comparison`
- Raw SKILL.md: https://api.skillmd.com/api/skills/samraddhashrivastavatech/comparison/raw
- Safety review: pending
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: Research & Search
- Author: SAMRADDHASHRIVASTAVATECH (https://skillmd.com/u/samraddhashrivastavatech)
- Updated: 2026-09-22
- Page: https://skillmd.com/skills/samraddhashrivastavatech/comparison

---


# Comparison

Drive structured comparison: entities across dimensions, or sources against
each other.

## When to use

- A/B/-C questions: "X vs Y vs Z" across cost, features, support, adoption.
- Choosing between products, libraries, approaches, or positions.
- Weighing two sources' reliability when they conflict.

## Tools used

- `compare_entities` — `entities` list, optional `dimensions`, `limit` per
  entity (default 4).
- `compare_sources` — directly analyze/compare source URLs.

## Workflow

1. Fix the entity set and the dimensions that matter (be explicit — the
   comparison is only as fair as its dimensions).
2. `compare_entities` (or `compare_sources` when you have the URLs).
3. Keep each entity's facts sourced; flag where evidence is missing.
4. Conclude per dimension, not with a single winner label unless supported.

## Input schema

```json
{ "entities": "list[str]", "dimensions": "list[str]", "limit": "int" }
{ "urls": "list[str]" }
```

## Output schema

```json
{ "matrix": "object", "notes": "list[str]" }
```

## Security

Comparisons invite selection bias; cover the same dimensions for every entity
and show the evidence for each cell.

## Related skills

`competitive-research`, `web-search`, `source-analysis`, `evidence-extraction`