joshua-zyy/academic-paper-writer / skills/academic-reviser
academic-reviser
VerifiedSelf-review, audit, or verify CS/AI/ML paper drafts as a critical peer reviewer. Three-round review (evidence→argument→style) with Verification Status and debt tracking. Use when: reviewing a paper draft before submission, checking evidence compliance of claims, simulating peer reviewer feedback, verifying citation closure and evidence debts, performing cross-section consistency checks. Triggers on: self review, 自查, verification, 审稿, evidence compliance, peer review, 论文审查, draft audit, 验证论文, 检查引用, cross-section review, 审修, draft verification.
| File | Kind | Size |
|---|---|---|
| SKILL.md | skill md | 3.4 KB |
| agents/reviser_agent.md | markdown | 5.0 KB |
| manifest.yaml | config | 1.9 KB |
| references/common-pitfalls.md | markdown | 1.4 KB |
| references/examples/reviser-output-examples.md | markdown | 3.0 KB |
| references/reviewer-package.md | markdown | 2.3 KB |
| references/revision-checklist.md | markdown | 5.1 KB |
| references/templates/section-critique.md | markdown | 1.1 KB |
| references/verification-checklists.md | markdown | 4.1 KB |
| references/verification-status.md | markdown | 4.0 KB |
| scripts/placeholder_audit.py | script | 3.6 KB |
| static/core/anti-patterns.md | markdown | 721 B |
| static/core/output-contract.md | markdown | 964 B |
| static/core/red-lines.md | markdown | 745 B |
| static/core/stance.md | markdown | 1.8 KB |
references/common-pitfalls.md
View on GitHub# 常见自欺模式
以下行为在审查和修订中要主动避免:
## 1. 只修措辞,不修证据
把"可能有效"改成"展现出潜力",但没有补实验或不实引文。
→ 措辞变更不改变证据等级。没有证据时,占位符优于文字包装。
## 2. 删除占位符却不补内容
把 `[REF_NEEDED: ...]` 删掉,让段落看起来完整,但没有真的找到引用。
→ 占位符是安全机制,不是需要隐藏的缺陷。补了内容再删。
## 3. 因长度而误信可信度
因为草稿已经很长(很多段落、很多引用),就假设它足够可信。
→ 长度 ≠ 证据链完整。逐一核验关键主张。
## 4. 因高指标而忽略协议缺陷
看到一个很高的指标(如 95% accuracy),就不再追问 split 是否可靠、有无泄漏。
→ 高指标不能证明协议正确。先审协议,再看数字。
## 5. 用华丽写法掩盖内容不足
用更漂亮的措辞和更长的段落掩盖 related work 只有两三个 cluster、Introduction 缺少领域缺口分析、Method 缺少设计论证。
→ 优雅的 prose 无法替代内容密度和论证深度。thin_draft 必须通过 Expansion Pass 解决。
## 6. 跳过检查顺序
先修风格再查事实,或者发现风格问题后不再回查证据。
→ 始终按"证据 → 论证 → 风格"顺序执行,发现前一层的遗漏必须回溯。
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| rea-analysis-optimizationmorluto/reaOptimize REA providers and shared analysis workflows for agent context, latency, CPU, and memory while preserving correctness and Evidence. Use for analysis bottlenecks and cache or query design; CI/test-suite optimization belongs to rea-ci-tests-audit. | Stars: 472k | Updated | Claude CodeCodex | Verified | Install |
| rea-ci-tests-auditmorluto/reaAudit REA CI and test suites for correctness, fixture drift, false passes, and measured performance; use for suite or workflow investigations and efficiency changes, rather than ordinary feature testing or a single test failure. | Stars: 472k | Updated | Claude CodeCodex | Verified | Install |
| rea-tool-designmorluto/reaAudit, investigate, design, or change REA investigation tools, CLI/MCP contracts, provider capabilities, and Evidence semantics. | Stars: 472k | Updated | Claude CodeCodex | Verified | Install |
| ponytail-auditDietrichGebert/ponytailQuality audit of the whole repo: bugs, security, real load, missing tests, speed, and what to delete. Most important first. | Stars: 196k | Updated | Claude CodeCodex | Verified | Install |