Hypothesis placement tournament
Skill panjose/Co-Scientist/skills/hypothesis-placement-tournament
Scientific agent skills for Claude Code and Codex that turn research goals into auditable hypothesis generation, review, ranking, evolution, and synthesis.
npx -y skills add panjose/Co-Scientist --skill hypothesis-placement-tournamentAssembled from the repository path, not quoted from the project. Check it against their README if it does not work.
One thing to look at
- 4 stars4 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.
What its author says it does
Copied from the file, not written here
Judge one placement-tournament matchup between a candidate hypothesis and one opponent.
SKILL.md
1.9 KB, as published. Nobody here has run it
hypothesis-placement-tournament
Goal:
- Judge one placement-tournament matchup between a candidate hypothesis and one opponent.
Inputs:
research_plan/RESEARCH_PLAN.json- two
hypotheses/<id>/HYPOTHESIS.jsonartifacts with completed review bundles
Outputs:
- one
TournamentMatchContractresult for a placement tournament
Context Loading:
- Open
skills/shared-references/schema-index.md. - Read
packages/agent_contracts/ranking.pyand confirm the exactTournamentMatchContractshape before emitting the placement result. - Read
research_plan/RESEARCH_PLAN.json. - Read both hypotheses and their review bundles.
- Confirm that the match strategy is
placement_tournament.
Execution Prompt Contract:
- System Intent:
- You are making one single-turn comparative judgment between two reviewed hypotheses.
- Required Reasoning Focus:
- Compare explanatory power, mechanistic quality, feasibility, and testability against the active research goal.
- Prefer the hypothesis that would be more valuable to keep in the frontier.
- Do Not Do:
- Do not return a tie.
- Do not ignore major review failures.
- Output Shape:
- Emit one
TournamentMatchContract. match_strategy:placement_tournamentwinner_id: one of the two hypothesis IDsreasoning: concise but specific
- Emit one
Execution Steps:
- Open
skills/shared-references/schema-index.md, then readpackages/agent_contracts/ranking.pybefore emitting the placement match artifact. - Compare the two reviewed hypotheses for frontier placement.
- Choose exactly one winner.
- Emit one valid placement-tournament match result.
Completion Rule:
- This skill is complete only when one valid placement match result has been produced with a non-empty winner and reasoning.