method-analyst

name	method-analyst
description	Use when you want Cursor Agent to turn method notes into structured formula cards as JSON under assets/notes/agent_artifacts/method/ — for your own editing workflow, not for any CLI tool.

Method analyst

This skill is for extracting method formula cards, not for writing the method section.

Use it when:

• the goal is to turn method materials into candidate formulas
• you will merge approved items into main.tex yourself (or with Agent help)
• the model should behave like an analyst, not an author

Do not use it to:

• draft the full Method section in one shot without human review
• approve formulas automatically
• write explanation outside the artifact file

Output contract

Write exactly one JSON artifact to:

assets/notes/agent_artifacts/method/formula_output.json

(create parent directories if needed). Read the schema in references/formula-output-schema.json before producing output.

Allowed top-level shapes:

{"formulas": []}

or a bare JSON array. If a field is unknown, leave it empty instead of inventing it.

Workflow

1. Read assets/notes/project.md, assets/notes/method.md (if present), and any assets/src/.
2. Read references/formula-output-schema.json.
3. Produce candidate formula cards only.
4. Save them to assets/notes/agent_artifacts/method/formula_output.json.
5. Stop.

Constraints

• Keep status as candidate.
• Keep allowed_sections limited to method and appendix unless the project context clearly requires less.
• Use citation_required conservatively.
• Mark is_core_formula as true only when the formula belongs in the main method narrative.
• Keep origin grounded in the available files.
• Keep symbols short and concrete.
• Do not include Markdown, comments, or prose in the artifact.

Prompts

For Codex, reuse references/codex-prompt.md.

For Claude, reuse references/claude-prompt.md.