Convert ALW and methodology outputs into executable pytest tests.
npx skills add https://github.com/gtylee/CodexGAS --skill test-writing
Please help me install this Agent Skill. Skill: Test Writing Source: https://github.com/gtylee/CodexGAS/tree/main/modelgas/skills/tests Command: npx skills add https://github.com/gtylee/CodexGAS --skill test-writing
This Skill translates risk assessments and validation requirements into concrete, executable tests that verify model behavior, numerical stability, and outcome reasonableness.
Request a test plan and pytest skeletons for the provided model risk artifacts, referencing the test_matrix.yaml under data/ and the ALW/methodology outputs.
High-intent search queries and answers about installing and using this skill.
This Skill generates pytest scripts from model risk assessments by converting ALW and methodology outputs into executable tests. It produces runnable pytest files that validate conceptual soundness, implementation correctness, and numerical stability.
Test traceability in financial model validation maps each generated test to specific ALW items and supporting evidence. This ensures clear evidence links and deterministic tests by specifying data schemas, seeds, tolerances, and traceability identifiers.
To create a test plan matrix for methodology outputs, request a test plan from the Skill referencing the test_matrix.yaml. It produces a matrix of test names, purposes, and categories tailored to your ALW artifacts.
Yes, you can specify data schema requirements for executable model tests. The Skill specifies schema-only dataset requirements and seeds when data is necessary for tests to ensure deterministic validation outcomes.
Yes, this approach works for numerical stability and outcome reasonableness checks. It applies executable tests across financial modeling workflows to validate implementation correctness and verify model risk assumptions.
gtyleeIdentify and categorize assumptions, limitations, and weaknesses in model outputs into a JSON schema.
LorenFrankLabImplement test-driven development for scientific code using pytest-based tests.
kirkchenInitialize Beat configuration and generate a minimal config.yaml.
jshsakuraMap AI workflow risks to prevention, detection, confirmation, and fallback guardrails.
