Validate a model claim by running a focused probe and merging evidence.
npx skills add https://github.com/dylan-conlin/orch-go --skill research-dylan-conlin
Please help me install this Agent Skill. Skill: research Source: https://github.com/dylan-conlin/orch-go/tree/main/skills/src/worker/research Command: npx skills add https://github.com/dylan-conlin/orch-go --skill research-dylan-conlin
This skill provides a disciplined, constrained approach to validating a single model claim by running one predefined probe method and recording concrete evidence, ensuring reproducibility and minimal drift in assessment.
Read SPAWN_CONTEXT.md and follow the constraint: pick exactly one claim and one method, run the probe, and merge the results into the model.
High-intent search queries and answers about installing and using this skill.
To validate a single model claim, read the spawn context, select exactly one predefined method, execute the probe, and write a parseable artifact. This structured approach ensures minimal drift and reproducible evidence collection.
Updating a knowledge base with probe evidence requires writing a probe artifact to the model directory and merging the finding into the parent model. This ensures traceability and governance for the observed evidence.
Recording concrete evidence when testing a model claim involves generating a parseable probe artifact in the designated probes directory. This evidence-first verification requires actual observations before merging results into the parent model.
No, this verification approach is strictly constrained to picking exactly one claim and one method per probe. This disciplined constraint prevents assessment drift and ensures reproducible evidence collection for the parent knowledge base.
Yes, reading the SPAWN_CONTEXT.md file is required before probing a model claim. This context provides the necessary constraints and scenario information to select one method and execute the validation probe correctly.
MMXCRecommend Proof of Life probe types from hypothesis, risk factors, and resource constraints.
gtyleeConsolidate validated findings from Skills 1–7 into docs/model_doc.md.
Best6668Refine vague modeling ideas into implementable proposals for math modeling contests.
KYRIE66nbExtract core claims and assess prior art across arXiv, Google Scholar, and conference databases.
