Import Theory

You are an expert scientific agent. Your goal is to import a pre-existing theory from a provided file (latex, pdf, or markdown) and rewrite it into a markdown file, while preserving as much of the content and structure as possible. You'll then store the theory into our context database for future use.

Input

Arguments: $ARGUMENTS

The arguments contain a path to the pre-existing theory file (with extension .tex, .pdf, or .md).

Folder setup

All commands must be run in the current working directory. Do not cd anywhere else, do not try to use the global /tmp folder or TMPDIR (only use the local ./tmp folder).

Set up an output folder: OUTPUT_DIR: mktemp -d -p ./tmp import-theory-output-XXXX

• <OUTPUT_DIR>/ — write the imported theory and all required image files here.

Any temporary files must be stored only under <OUTPUT_DIR>.

Reading the input theory

• .md / .tex: read with the Read tool directly.
• .pdf: read with the Read tool (it handles PDFs natively). For large PDFs (>10 pages), read page ranges incrementally with the pages parameter.

Copy and/or extract all images and figures from the source into <OUTPUT_DIR>/ so you can reference them in your final markdown.

Execution Steps

1. Parse input: Extract the file path from $ARGUMENTS. Validate the file exists and has extension .tex, .pdf, or .md.
2. Extract figures: If the source file contains images/figures, extract and save and/or copy them into <OUTPUT_DIR>/ and note their new paths for referencing in the markdown. Crop the relevant sections of a PDF to isolate figures if needed. Inspect each image file to verify that you have obtained the correctg content.
3. Convert file: Write the theory as a markdown file to <OUTPUT_DIR>/theory.md (this exact filename is required). Make sure you maintain any mathematical formulas. Correctly reference any images relative to <OUTPUT_DIR>.
4. Store results: Persist your output and return the theory ID:

   uv run python <SKILL_BASE_DIR>/scripts/context_manager.py store_results --from_agent_type import-theory --from_folder <OUTPUT_DIR>
   

   Note down the returned theory ID (e.g. T_20260421_150000_x1y2z3) as the result of this skill.