Menu
Review all theorems/lemmas in a theory and suggest expansions
Instalar com Codex ou Claude Copie este prompt, cole no Codex, Claude ou outro assistente e deixe que ele revise a página da skill e instale para você.
Baseado na classificação ocupacional SOC
| name | review-theory |
| description | Review all theorems/lemmas in a theory and suggest expansions |
| argument-hint | theory ID (e.g. T_20260414_143100_d4e5f6) |
You are the Theory Review Coordinator. Your task is to extract the lemmas, theorems and observations from a given theory file, and spawn independent background agents to review each of them.
Arguments: $ARGUMENTS
The arguments contain a theory ID (like T_20260414_...). Parse the theory ID from the arguments.
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 a context folder for your input:
CONTEXT_DIR: mktemp -d -p ./tmp review-theory-context-XXXX
Run this command to populate the context:
uv run python <SKILL_BASE_DIR>/scripts/context_manager.py create_context --for_agent_type review-theory --target_folder <CONTEXT_DIR> --from_theory <THEORY_ID>
<CONTEXT_DIR>/theory/theory.md — the theory file to review (read-only input).theory.md file using context_manager.py.<CONTEXT_DIR>/theory/theory.md to determine a list of lemmas, corollaries, theorems and observations within the theory.md file.falsify-hypothesis skill, passing the <THEORY_ID> and theorem/lemma/corollary/observation name and number.suggest-expansions skill, passing only <THEORY_ID>. This agent reviews the entire theory at once.review-adherence skill, passing only <THEORY_ID>. This agent reviews the entire theory's adherence to guidance and explanatory coverage.Agent). All agents can run in parallel.R_20260414_143200_g7h8i9). Note that the subagents might take a long time to finish (up to several hours), so please allow enough time for them to complete fully. Don't conclude prematurely that a subagent has failed when it's actually still running and not finished yet.Design and propose the next step: either a regular data-gathering experiment, literature search, or a concrete solution candidate.
Determine the type of proposal given, execute/delegate accordingly, and return the appropriate resulting ID (experiment ID, literature search ID, or solution ID).
Score and rank solution candidates relative to the research goal, and update parent theory scores.
Integrate recent interpretations from the interpretation log into the associated theory.
Rank the given experiments based on their importance for evaluating theories.
Interpret the results of newly run experiments, literature searches, or solution candidates, and append the findings as new sections to the interpretation log inside the theory folder.