Test design, vulnerability assessment, and quality assurance (13K skills)
Use when building the robustness section for a 《会计研究》 (Accounting Research) manuscript — replacing accounting measures with alternatives, adding governance / institutional controls, placebo tests, and endogeneity follow-ups — so the main result survives an accounting referee's scrutiny. Use after mechanism is settled.
The English summary is being prepared. Showing the original description.
Use when strengthening reproducibility evidence for an ACL paper reviewed through ACL Rolling Review, covering the Responsible NLP checklist end to end, hyperparameter and compute reporting, prompt and decoding disclosure for LLM experiments, data contamination auditing, variance across runs, and checklist-to-paper consistency.
The English summary is being prepared. Showing the original description.
Use when the empirical identification strategy is the bottleneck for a 《会计研究》 (Accounting Research) manuscript — exogenous standard / regulatory changes (event study, DID), PSM, Heckman, IV — and you need to stress-test the design before drafting tables. Prefers standard/regulatory shocks as the cleanest accounting identification source.
The English summary is being prepared. Showing the original description.
Use when the mechanism analysis for a 《会计研究》 (Accounting Research) manuscript is missing or mis-specified — the channel must be an information mechanism (reduce information asymmetry, raise disclosure quality, constrain earnings management, improve audit assurance), not a generic pricing or financing channel. Use after the main result holds.
The English summary is being prepared. Showing the original description.
Use when choosing and applying a theory-construction method for an Academy of Management Review (AMR) manuscript — building constructs, defining their domain, specifying relationships, articulating the underlying logic/mechanisms, and setting boundary conditions. This is the THEORY-CONSTRUCTION craft, NOT empirical method; AMR publishes no datasets, no measures, and no statistical tests.
The English summary is being prepared. Showing the original description.
Use when designing or auditing experiments for an ACL paper, covering tuned LLM baselines, multi-dataset and multilingual evaluation, statistical significance and variance, human evaluation with agreement reporting, contamination and prompt-sensitivity controls, ablations, and error-analysis expectations in NLP reviewing.
The English summary is being prepared. Showing the original description.
Use when choosing or sharpening a research question for an Accounting, Organizations and Society (AOS) manuscript — testing whether the question genuinely concerns the social, organizational, behavioral or institutional life of accounting, and picking the intellectual tradition the question demands. Screens fit; it does not build the theory (aos-theory-development).
The English summary is being prepared. Showing the original description.
Use when building the exhibits of an Accounting, Organizations and Society (AOS) manuscript — data-inventory and evidence tables for qualitative work, cell-means and process-test tables for experiments, and figures that carry theoretical weight — in Elsevier-compatible format. Builds exhibits; it does not run the analysis (aos-data-analysis).
The English summary is being prepared. Showing the original description.
Use when packaging ACM CCS artifacts for the artifact-evaluation committee and the ACM badges — Artifacts Available, Artifacts Evaluated Functional, Artifacts Evaluated Reusable, and Results Reproduced — covering what security evaluators inspect, how to make attacks and defenses turnkey, and how to justify withheld artifacts.
The English summary is being prepared. Showing the original description.
Use when writing the contribution and implications section for a 《会计研究》 (Accounting Research) manuscript — articulating both the accounting-theory contribution (role of accounting information) and concrete implications for standard-setters, regulators, and audit/reporting practice. Use after results and tables are final.
The English summary is being prepared. Showing the original description.
Use when designing or auditing AAMAS experiments - self-play and population-based training, opponent selection, equilibrium and regret metrics, game-theoretic simulations, ablations, seeds, hyperparameters, compute, and claim-to-evidence fit - with emphasis on experiments that probe the interaction rather than chase a single-agent leaderboard.
The English summary is being prepared. Showing the original description.
Use when auditing an AAAI main technical track submission for OpenReview readiness, double-blind anonymity, page limits, reproducibility checklist, supplementary material, author limits, multiple-submission policy, and AAAI AI-use policy compliance.
The English summary is being prepared. Showing the original description.
Use when auditing an AAMAS main-track submission for OpenReview readiness, the 8-page-plus-references limit, the full-paper versus 2-page extended-abstract choice, double-blind anonymity, the supplementary zip cap, track fit across the AAAI/JAAMAS/Blue Sky tracks, dual-submission rules, and desk-reject triggers before upload.
The English summary is being prepared. Showing the original description.
Use when appraising the cumulative evidence and ensuring balance in an Academy of Management Annals (Annals) review — weighing conflicting findings by credibility, steelmanning rival schools, and handling the author's own work even-handedly. Audits evidence quality and fairness; it does not design the framework (amann-organizing-framework) or build exhibits (amann-tables-figures).
The English summary is being prepared. Showing the original description.
Use when designing or auditing AAAI experiments for the broad-AI program committee, including baselines, ablations, statistical significance, robustness, human evaluation, AI-for-Social-Impact and alignment/safety evidence, compute and cost reporting, and reproducibility-checklist alignment for Phase-1 survival.
The English summary is being prepared. Showing the original description.
Set a repo up to prove engineering-task work actually works before it ships. Investigates the repo, ensures a one-command dev stack (`dev-local`) exists, asks whether verification runs locally or in a sandbox (crabbox), confirms/installs the driver (the `playwright-cli` skill for web by default). Outputs three artifacts: a committed `/verify` skill (per-task verification SOP — spawn a verifier sub-agent → drive the app → screenshot/video proof → open a PR with the proof embedded), the `/dev-local` skill + script, and the installed driver skill. Use when someone says "set up verification", "make this repo verifiable", "scaffold a verify skill", "set up the verifier".
The English summary is being prepared. Showing the original description.
Audit a repo's agent context — CLAUDE.md files, codebase docs, skills, and tool/MCP designs — against Anthropic's Claude 5 context-engineering guidance ("unhobbling": Anthropic cut ~80% of Claude Code's system prompt with no eval loss). Finds overconstraint, conflicting instructions, redundancy, stale facts, and missing "unknown knowns"; produces a scored findings report with concrete rewrites, then applies approved fixes. Use when someone says "audit my CLAUDE.md", "context audit", "unhobble this repo", "review our agent docs/skills/tools", or after upgrading to Claude 5-generation models.
The English summary is being prepared. Showing the original description.
Set up an end-to-end test suite in any repo, following practices that make e2e a reliable per-PR gate: real flows over bypass, layered assertions, a reusable auth/session helper, video+trace evidence, and a compounding suite. Use when a repo has no e2e (or weak e2e) and you want system-level tests — "set up e2e", "add end-to-end tests", "scaffold a test gate".
The English summary is being prepared. Showing the original description.
Spin up a new loop (domain) in a file-based knowledge base — bootstrap the substrate if it's missing, gather the loop's charter, scaffold domains/<loop>/README.md, then do ONE real test run and record it in the loop's Timeline and LOG.md. Use when the user says "set up a new loop", "create a domain", "start a new beat/workstream", or names a recurring job they want the agent to own.
The English summary is being prepared. Showing the original description.
Scaffold an isolated CLOUD dev box per agent (via crabbox + Daytona) for any codebase — the parallel-safe counterpart to dev-local-setup. Each agent gets its own full stack (own DB + dev server) and an in-box browser for e2e, so concurrent loops never collide on ports/state. Sets up the snapshot image, .crabbox.yaml, an idempotent setup.sh (also boots the stack locally), and a cbx.sh wrapper. Use when the user says "set up crabbox", "give each agent its own box", "add cloud testing", "make this repo testable in the cloud / on Daytona", "parallel-test this", or when setup-codebase-harness needs true per-agent isolation.
The English summary is being prepared. Showing the original description.
Use when wiring an existing DataFusion or datafusion-spark function into Comet for a Spark expression. Identifies the right wiring pattern (one-line passthrough, datafusion-spark UDF registration, or custom serde with input massaging / restrictions), applies the Scala serde, registers the UDF in jni_api when needed, and adds SQL file tests. Assumes the function already exists upstream — if not, switch to `implement-comet-expression`.
The English summary is being prepared. Showing the original description.
Audit an existing Comet expression for correctness and test coverage. Studies the Spark implementation across versions 3.4.3, 3.5.8, 4.0.1, and 4.1.1, reviews the Comet and DataFusion implementations, identifies missing test coverage, and offers to implement additional tests.
The English summary is being prepared. Showing the original description.
Use when implementing a new Spark expression in DataFusion Comet. Walks through cloning latest Spark master to study the canonical implementation, checking the upstream datafusion-spark crate before writing native code, building the Comet serde and Rust wire-up from the contributor guide, then running audit-comet-expression to drive a test-coverage iteration loop.
The English summary is being prepared. Showing the original description.
Use when picking the next Spark expression to implement natively in Comet, or when asked whether a specific codegen-dispatched expression is worth a native Rust implementation. Scores the candidate on how likely a native path can be 100% Spark-compatible and how much throughput or allocation it would save versus the JVM codegen dispatcher, records the verdict in the per-expression audit log so disqualified candidates are not re-litigated, and files a GitHub issue for the recommendation.
The English summary is being prepared. Showing the original description.