Results for “differential-review”

通用多轮迭代改进：对任意研究制品反复调用 /review → 解析反馈 → 修复 → 更新 wiki，直到达标

Launches four parallel read-only sub-agents to review a git diff or file scope for regressions, security risks, performance issues, and contract gaps, then aggregates findings into a prioritized fix path.

Use after Cell reviews arrive to triage the decision, prioritize the (often substantial) new experiments by impact × feasibility, and draft a point-by-point response that is respectful, evidence-led, and closes the "completeness" gaps reviewers demand. Do not run before the experiments and revision are actually done.

Econometrics skill for Difference-in-Differences (DID) analysis. Activates when the user asks about: "difference in differences", "DID", "DiD", "diff-in-diff", "parallel trends", "treatment group", "control group", "pre-treatment", "post-treatment", "policy evaluation", "natural experiment", "staggered DID", "event study regression", "two-way fixed effects DID", "callaway santanna", "sun and abraham", "双重差分", "倍差法", "平行趋势", "处理组", "对照组", "政策评估", "事件研究", "交错DID", "渐进处理"

Use when implementing risk-adjusted rewards, discussing Sharpe ratio in RL training, or tuning reward components for risk awareness

Multi-perspective dialectical reasoning with cross-evaluative synthesis. Spawns parallel evaluative lenses (STRUCTURAL, EVIDENTIAL, SCOPE, ADVERSARIAL, PRAGMATIC) that critique thesis AND critique each other's critiques, producing N-squared evaluation matrix before recursive aggregation. Triggers on /critique, /dialectic, /crosseval, requests for thorough analysis, stress-testing arguments, or finding weaknesses. Implements Hegelian refinement enhanced with interleaved multi-domain evaluation and convergent synthesis.

Reviews a PR or diff through three independent lenses—correctness, security, and test coverage—then deduplicates and validates findings before persisting them, reducing noise while catching more real bugs.

Reviews code diffs or explicit file sets through independent correctness and QA lenses, producing a severity-ordered verdict for pre-merge audits or review gates.

Use when reviewing a branch, PR, or work-in-progress changes against standards and spec

Routes quality reviews to specialized critic agents based on file type or flags, covering peer review, code review, and manuscript polish.

Use after PNAS reviews arrive to triage the decision, prioritize experiments, and draft a point-by-point response that is respectful, evidence-led, and honest about limits. Do not run before the main text is actually revised.

doubao-contract-reviewer 是面向大众用户的合同审查 Skill，适合在豆包/豆包 Turbo 中审查各类合同。用户上传合同或询问“帮我审合同、合同有没有问题、这份合同能不能签、合同风险、合同把关、legal review”时必须使用。本 Skill 强制输出比裸跑更有用的结构化审查：先判断我方立场，再按交易模块识别风险，区分“必改风险 / 可争取优化项 / 形式完善项”，并给出可直接替换或补充的修改文本。适用于保密、买卖、服务、委托、借调、SaaS、许可、合作、租赁、渠道、数据处理等泛合同场景。

Reviews code using pi CLI with read-only tools. Delegates the review to pi (dev/pi) with a structured review rubric, running in read-only mode to prevent accidental edits. By default reviews uncommitted working tree changes (git diff HEAD) with pi restricted to the read tool; explicit targets (--branch/--diff/@file/PR) or --explore widen pi to read,grep,find,ls. Use when the user asks to "review code with pi", "pi review", "have pi review", "let pi review", or invokes /pi:review.

Runs a structured code review using four parallel subagents covering style, correctness, security, and performance, then synthesizes findings by severity and gives an approve/needs-changes recommendation.

Simulates a NeurIPS/SC/ICSE-style peer review with five reviewer personas (HPC, ML, Stats, Reproducibility, Devil's Advocate) that verify every claim against actual result data before submission.

Dispatch a code reviewer subagent to catch issues before they cascade, with precisely crafted context for evaluation.

Runs the current git diff through the Codex CLI for an independent, read-only second-opinion review, then triages findings back into the session for Claude to apply fixes.

Reviews pull request diffs for correctness, security, maintainability, and test coverage, providing severity-ranked findings and a close-out recommendation.

Deprecated standalone review skill