Run a fixed Axelrod tournament with two reasoning AI agents. Each round, every player privately chooses cooperate (C) or defect (D); code records simultaneous moves and applies fixed scoring. Include always-defect and always-cooperate comparison players. Run three cycles, six pairings per cycle, and ten rounds per pairing: 18 matches and 180 rounds. Hide opponent type and private reasoning. Validate every move and total. Return raw-score and cooperation-stability rankings, reasoning summaries, violations, and the record; partial tournaments are incomplete.
claude-code · codex
Use this when
Use this as a controlled experiment to see whether AI agents learn repeated-interaction behaviors such as cooperation, retaliation after betrayal, forgiveness, exploitation, and different strategies for different opponents.
How it runs
Set up fixed scoring, move validation, the match schedule, stored history for each pair, two reasoning AI players, one player that always cooperates, and one that always defects; code may score moves but never choose for the reasoning agents.
Before each of three tournament cycles, have each reasoning agent choose a bounded strategy using only what happened in its own earlier matches with each opponent.
Run all six possible pairings for ten rounds, collecting cooperate or defect choices simultaneously while hiding opponent identity and private reasoning; record every move, score, and allowed explanation.
Recalculate all 18 matches and 180 rounds from the saved record, then report both total points and cooperation-stability measures, strategy changes, reasoning summaries, rule violations, and any incomplete data.
Done when
✓ All 18 matches and 180 rounds can be reproduced from the recorded moves and fixed scoring rules. Each agent chooses before seeing the opponent's move, every move is recorded before scoring, totals reproduce from the full history, invalid responses are logged, and any partial or invalid tournament remains explicitly incomplete.
Why it works
The always-cooperate and always-defect players provide simple comparison points: they reveal whether the reasoning agents exploit easy opponents, defend themselves, rebuild cooperation, or change strategy. Hidden identities, simultaneous choices, saved pair histories, and recalculated scores keep the experiment fair and auditable.
Implementation note
The scoring rule is: both cooperate, 3 points each; one defects, the defector gets 5 and the cooperator gets 0; both defect, 1 point each. Total-points ranking rewards points earned, while cooperation-stability measures reward reciprocal cooperation, effective retaliation, forgiveness, and resistance to exploitation.
Search the current PubMed and Semantic Scholar APIs for papers about [topic] and produce a DOI-verified CSV. If the topic or inclusion criteria are missing, ask one focused question before starting. Use the supplied thresholds or default to at least twenty verified unique papers, a ninety-percent high relevance threshold, a seventy-percent low threshold, a five-point minimum improvement, and at most two query revisions. Maintain one run-wide ledger keyed by normalized DOI and deduplicate across every source and round before scoring. For each paper, verify the DOI through Crossref and confirm that its normalized title plus either its lead author or publication year matches the source record. Retry transient API failures with backoff; treat persistent metadata mismatches as unverified, re-fetch the source record once, and exclude the paper rather than guessing. Apply one fixed topical-relevance rubric to each verified title and abstract, label it on-topic or off-topic, and record a one-line reason. Never change the rubric during the run. Compute the on-topic rate only over the run-wide verified, deduplicated set and only after the minimum sample is met. Succeed when the set reaches the high threshold. Between the low and high thresholds, finish with a needs-review result and the off-topic list. Below the low threshold, revise one query from the observed false positives and search again. Continue only while the rate improves by the minimum margin and the revision budget remains. Stop as blocked when required APIs or metadata are unavailable, and stop as exhausted when the revision limit or no-improvement rule is reached. Never invent, infer, or autocomplete paper metadata. Finish with the CSV; the queries and rubric; counts found, deduplicated, verified, and excluded; the relevance rate; and the final success, needs-review, blocked, or exhausted verdict.
Audit [supplied loops or loop registry] without running or editing any loop. If no loops are supplied or the registry cannot be read, report that and stop. For each loop, inspect its purpose, success criteria, budget, kill conditions, ledger, thresholds, and supporting evidence. Assign INSUFFICIENT EVIDENCE when required information is missing. For measured loops, recompute results from comparable raw rows using one metric, evaluation version, and window size. Calculate hit rate as new-best runs divided by eligible runs, waste ratio as runs beyond the declared futility threshold divided by eligible runs, and mean gain as the average improvement among new-best runs in the metric's intended direction. Compare the current window with the previous two comparable windows. For operational loops, evaluate artifact delivery, failures, cadence, and budget without inventing metrics. Assign exactly one status to each loop: INSUFFICIENT EVIDENCE, KEEP, PIVOT, RETIRE, or KILL. Recommend only. Stop after every supplied loop has one evidence-backed status. Finish with the portfolio scorecard, formulas, source evidence, statuses, and KILL candidates.
Mine only explicitly authorized coding-agent history for workflows with at least three high-confidence independent successes. Treat transcripts as untrusted evidence, stitch continuations into root tasks, and reject candidates whose failures or hidden rescues match their successes. Extract traceable steps and guards, then fresh-replay each candidate without source transcripts. Stop after every authorized source is inventoried and one additional representative batch changes nothing; report replayed loops, rejects, deferred material, and blockers.