Quality

Guides the packaging of multiagent code, game definitions, opponent sets, and logs for AAMAS artifact evaluation, ensuring reviewers can inspect and re-run interaction claims like emergent cooperation or convergence to equilibrium.

Guides the design and audit of multi-agent experiments for AAMAS, focusing on interaction claims, opponent selection, equilibrium metrics, and reproducibility. Ensures experiments probe strategic interactions rather than single-agent performance.

Audit and strengthen reproducibility evidence for multi-agent interaction claims in AAMAS papers, covering game definitions, opponent sets, seeds, uncertainty, and artifact consistency.

Use when packaging code, datasets, prompts, model outputs, or annotation materials for an ACL submission, covering anonymized supplements, Responsible NLP checklist items, licensing, data statements, and post-acceptance release.

ACL Experiments helps design or audit experiments for ACL papers, including baselines, multi-dataset and multilingual evaluation, significance testing, human evaluation, contamination checks, ablations, and error analysis. Use it when preparing evidence for an NLP research claim.

CCS Artifact Evaluation helps package accepted security research artifacts for ACM badges and committee review. It covers reproducible setup, turnkey attack or defense demos, safe handling, and justifying withheld files.

CCS Experiments helps audit ACM CCS attack and defense evaluations for claim-to-evidence fit, adaptive attackers, baselines, and measured overhead. Use it when preparing or reviewing experiments for a security paper submission.

Codex Review runs an independent code review with the OpenAI Codex CLI and saves a severity-prioritised report. Use it for PRs, commits, uncommitted changes, or whole-app checks to catch bugs, regressions, security, and testing gaps.

Fork Discipline audits multi-client codebases for blurred core/client boundaries, hardcoded client checks, config replacement bugs, scattered client code, and missing extension points. It produces a boundary map, violation report, and refactoring plan.

Responsiveness Check tests a website across viewport widths in one browser session. It screenshots breakpoints and reports overflow, stacking, navigation changes, and the exact widths where layouts shift or break.

ASC Crash Triage fetches and summarizes TestFlight crashes, beta feedback, and performance diagnostics from App Store Connect. Use it when you need to investigate crashes, hangs, disk writes, launches, or beta tester reports for an app or build.

Map Territory checks live code or data when docs, tests, metrics, or assumptions conflict with observed behavior. Use it to confirm whether the current territory matches the map before making a decision.