Claude Skills by gtynnn060110-hash

Guidance on reading, parsing, and processing various file formats (CSV, JSON, YAML) commonly used in legal and compliance tasks

Update or correct an existing skill file based on judge feedback or improved understanding.

Create a new reusable skill file to capture a calculation method, formula, or tool pattern you've discovered during a task.

Guidance on reading, parsing, and processing various file formats (CSV, JSON, YAML) commonly used in legal and compliance tasks

Update or correct an existing skill file based on judge feedback or improved understanding.

Create a new reusable skill file to capture a calculation method, formula, or tool pattern you've discovered during a task.

Guidance on reading, parsing, and processing various file formats (CSV, JSON, YAML) commonly used in legal and compliance tasks

Update or correct an existing skill file based on judge feedback or improved understanding.

Create a new reusable skill file to capture a calculation method, formula, or tool pattern you've discovered during a task.

Guidance on reading, parsing, and processing various file formats (CSV, JSON, YAML) commonly used in legal and compliance tasks

Update or correct an existing skill file based on judge feedback or improved understanding.

Create a new reusable skill file to capture a calculation method, formula, or tool pattern you've discovered during a task.

Guidance on reading, parsing, and processing various file formats (CSV, JSON, YAML) commonly used in legal and compliance tasks

Update or correct an existing skill file based on judge feedback or improved understanding.

Create a new reusable skill file to capture a calculation method, formula, or tool pattern you've discovered during a task.

Guidance on reading, parsing, and processing various file formats (CSV, JSON, YAML) commonly used in legal and compliance tasks

Update or correct an existing skill file based on judge feedback or improved understanding.

Create a new reusable skill file to capture a calculation method, formula, or tool pattern you've discovered during a task.

Guidance on reading, parsing, and processing various file formats (CSV, JSON, YAML) commonly used in legal and compliance tasks

Update or correct an existing skill file based on judge feedback or improved understanding.

Create a new reusable skill file to capture a calculation method, formula, or tool pattern you've discovered during a task.

Guidance on reading, parsing, and processing various file formats (CSV, JSON, YAML) commonly used in legal and compliance tasks

Update or correct an existing skill file based on judge feedback or improved understanding.

Materials science toolkit. Crystal structures (CIF, POSCAR), phase diagrams, band structure, DOS, Materials Project integration, format conversion, for computational materials science.

Use this skill for LAB-Bench and biomedical question answering. 1. Treat candidate answers as controlled options. 2. Compare each option against the wording of the question and any source metadata. 3. Avoid adding synonyms in exact-answer tasks; return the exact option string. 4. If evidence is incomplete, state the uncertainty in `reasoning` while keeping `answer` concise.

Create a new reusable skill file to capture a calculation method, formula, or tool pattern you've discovered during a task.

Guidance on reading, parsing, and processing various file formats (CSV, JSON, YAML) commonly used in legal and compliance tasks

Use this skill for SciBench physics and quantitative science calculation tasks. 1. Extract variables, constants, units, and target quantity. 2. Choose the governing equation or theorem before substituting numbers. 3. Use `sympy`, `scipy`, or a small Python script for algebra or numerical checks when useful. 4. Keep unit conversions visible. If the verifier expects a numeric answer, put only the numeric value in `answer`. 5. Include the derivation summary in `reasoning`.

Update or correct an existing skill file based on judge feedback or improved understanding.

Use this skill before solving any science task. 1. Restate the scientific question, quantities, units, and requested output. 2. Identify whether the task is numeric, multiple choice, data analysis, or open-ended workflow design. 3. Read `/root/data/task-XXX/task_payload.json` before making assumptions. 4. Track assumptions explicitly and prefer dimensional checks for quantitative answers. 5. Write a compact final JSON file to the exact `/app/task_XXX_output.json` path.

Use this skill for tasks that mention datasets, previews, CSV files, models, or generated artifacts. 1. Inspect the payload, folder tree, preview rows, and requested output paths. 2. Propose a reproducible script layout before modeling. 3. Choose baselines and metrics appropriate to the domain. 4. Validate by checking file existence, schema, row counts, value ranges, and leakage risks. 5. Save generated artifacts under `/app` unless the task explicitly asks for another path.

Use this skill when a task points to external scientific sources or benchmark pages. 1. Prefer primary benchmark URLs and cited dataset/project pages. 2. Record accessed URLs and version assumptions in the output JSON. 3. Distinguish source facts from derived calculations. 4. Do not claim unavailable files were inspected; describe fallback reasoning instead.

Guidance on web content analysis, data extraction, and online research techniques for legal and regulatory information gathering