Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add ustc-ai4science/academic-search --skill academic-searchgit clone --depth 1 https://github.com/ustc-ai4science/academic-searchWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/ustc-ai4science/academic-search/academic-search)<a href="https://agentmods.dev/skills/ustc-ai4science/academic-search/academic-search"><img src="https://agentmods.dev/badge/skills/ustc-ai4science/academic-search/academic-search/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/ustc-ai4science/academic-search/academic-search"><img src="https://agentmods.dev/badge/skills/ustc-ai4science/academic-search/academic-search.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00088 | $0.01995 |
| Opus 5.5 | $0.00035 | $0.00798 |
| Sonnet 5.5 | $0.00018 | $0.00399 |
| Haiku 4.5 | $0.00009 | $0.00199 |
Grade A, and why
academic-search scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2mo ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
Copies of this mod
2 near-identical copies found in the catalogue:
- academic-search — 100% identical, 424 lines differ
- pr-academic-search — 97% identical, 426 lines differ
How it starts
The opening of the file, as written. The whole thing — 60 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Academic Search
按用户要求检索、筛选并交付可追溯的学术资料。优先官方 API 和论文来源；浏览器用于动态页面、交互或当前 API 无法满足的字段。
任务范围与完成标准
从请求确定学科、论文范围、年份、数量和所需字段；只有影响结果且无法合理推断的缺项才询问。开放式选题可先轻量检索、筛选，再补充核心论文；用户已经要求完整元数据、摘要、BibTeX 或下载时，直接完成相应步骤，不因“两遍策略”再次要求确认。
完成标准是所需记录、字段和文件均已交付，或逐项说明未完成部分的证据与原因。工具返回成功、页面加载完成、URL 看起来像 PDF 都不能单独证明任务完成。
学科路由与来源
读取与任务对应的学科 profile，跨学科任务以核心学科为主，并用 OpenAlex / Crossref 等补充。
| 学科 / 任务 | 优先来源 | 按需参考 |
|---|---|---|
| 计算机 / AI | arXiv、Semantic Scholar、ACM DL、IEEE、DBLP | computer-science |
| 医学 / 生命科学 | PubMed、PMC、Europe PMC、临床试验库 | biomedicine |
| 物理 / 数学 | arXiv、NASA ADS、INSPIRE HEP | physics-math |
| 化学 / 材料 | Crossref、OpenAlex、ChemRxiv、出版商 | chemistry-materials |
| 经济 / 社科 | RePEc、NBER、SSRN、OSF | economics-social-science |
| 人文 / 法律 | 学术搜索、图书馆目录、领域数据库 | humanities-law |
| 精确论文 / 元数据 | DOI → Crossref / 出版商；arXiv ID → arXiv；S2 / OpenAlex 补全 | API cookbook |
| 引用 / 作者论文 | S2 API；用户指定 Scholar 时使用浏览器 | S2、Scholar |
| 中文文献 | CNKI 等中文数据库，按当前访问条件检索 | CNKI |
| OA PDF / 全文 | 论文来源、开放仓储、OpenAlex、Unpaywall、PMC | 全文工作流 |
只使用 API 时不启动浏览器或执行 CDP 环境检查。调用参数、鉴权与分页见 API cookbook。代码链接优先核对论文或作者给出的仓库；第三方索引作为发现线索。
检索与核验
- 根据研究问题选择互补查询词、受控词表与文献类型。记录实际查询、筛选条件和检索时间；按相关性及用户目标排序。CCF 只用于适用的 CS 任务，其他学科遵循 profile；分级需注明来源和版本。
- 轻量筛选可暂不拉完整摘要；完整元数据任务应继续补齐已要求的字段。论文表明确区分搜索片段、摘要与全文证据。
- 按 metadata-schema 输出。需要交付或复用 JSON 记录时，按 academic-records 运行可执行校验与保守去重：DOI/arXiv/PMID 等精确标识可归组，规范化后标题完全相同只产生待核对候选。预印本与正式版本保留对应关系，勿静默合并不同研究。
- 引用数保留平台、数值、检索时间和来源；不同平台的计数分别保存，不求和、不把较大值当唯一真值。缺失字段用空值及原因，不补造。
- API 429 只说明当前请求被限流，不能确定会话配额永久耗尽。遵守有效的
Retry-After，无提示时有界退避；同一请求默认最多 2 次重试、总等待预算 30 秒，预算不足则停止该来源并记录限制。并行任务共享平台请求预算；参数详见 API cookbook 与 S2 经验。
What ships with it
52 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
- agents/openai.yaml 354 B
- assets/academic-search-framework.png 8101 KB
- assets/logo.png 3062 KB
- references/academic-records.md 7.4 KB
- references/api-cookbook.md 13 KB
- references/browser-workflow.md 8.2 KB
- references/cdp-api.md 23 KB
- references/disciplines/biomedicine.md 1.4 KB
- references/disciplines/chemistry-materials.md 1.2 KB
- references/disciplines/computer-science.md 1.3 KB
- references/disciplines/economics-social-science.md 1.2 KB
- references/disciplines/humanities-law.md 1.4 KB
- references/disciplines/physics-math.md 1.1 KB
- references/full-text-workflow.md 4.9 KB
- references/metadata-schema.md 18 KB
- references/rankings/biomed-evidence-ranking.md 1.6 KB
- references/site-patterns/arxiv.org.md 2.1 KB
- references/site-patterns/cnki.net.md 4.7 KB
- references/site-patterns/dl.acm.org.md 2.5 KB
- references/site-patterns/ieeexplore.ieee.org.md 2.6 KB
- references/site-patterns/link.springer.com.md 1.1 KB
- references/site-patterns/onlinelibrary.wiley.com.md 1.2 KB
- references/site-patterns/paperswithcode.com.md 2.8 KB
- references/site-patterns/pubmed.ncbi.nlm.nih.gov.md 3.3 KB
- references/site-patterns/pubs.acs.org.md 1.1 KB
- references/site-patterns/scholar.google.com.md 3.9 KB
- references/site-patterns/sciencedirect.com.md 1.3 KB
- references/site-patterns/semanticscholar.org.md 4.1 KB
- references/venue-rankings.md 5.1 KB
- references/workflows/systematic-review.md 3.1 KB
- scripts/academic-records.mjs 41 KB runs code
- scripts/academic-records.test.mjs 32 KB runs code
- scripts/browser-cdp.test.mjs 36 KB runs code
- scripts/browser-page.mjs 37 KB runs code
- scripts/browser-page.test.mjs 30 KB runs code
- scripts/browser-runtime.mjs 11 KB runs code
- scripts/browser-runtime.test.mjs 7.9 KB runs code
- scripts/browser-smoke.mjs 16 KB runs code
- scripts/browser-test-isolation.test.mjs 1.6 KB runs code
- scripts/cdp-proxy.mjs 34 KB runs code
- scripts/check-deps.sh 841 B runs code
- scripts/ensure-proxy.mjs 6.3 KB runs code
- scripts/ensure-proxy.test.mjs 1.8 KB runs code
- scripts/fixtures/pdf/corrupt-page-tree.pdf 603 B
- scripts/fixtures/pdf/valid.pdf 602 B
- scripts/oa-pdf-download-self-test.sh 4.1 KB runs code
- scripts/oa-pdf-download.mjs 11 KB runs code
- scripts/oa-pdf-download.test.mjs 13 KB runs code
- scripts/proxy-owner.mjs 3.1 KB runs code
- scripts/proxy-owner.test.mjs 3.8 KB runs code
- scripts/release-test.sh 14 KB runs code
- scripts/self-test.sh 12 KB runs code
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 2mo ago Changed · +15 tokens per session eca9045b18da
- 2mo ago Changed · -348 lines · -63 tokens per session scan D → A 1ab9c8d0ff6f
- 2mo ago First seen · 408 lines · 136 tokens per session scan D 7848d2c1318f
academic-search is a skill published in the GitHub repository ustc-ai4science/academic-search (704 stars, last pushed 2mo ago), licensed MIT. It adds 88 tokens to every session and 1,995 once invoked, about $0.0004 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
academic-search compared with skills from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| academic-searchustc-ai4science/academic-search | Skill | 704 | 2mo ago | MIT | A | 88 | Codex |
| imagegenopenai/codex | Skill | 133k | yesterday | Apache-2.0 | A | 113 | Codex |
| playwright-climicrosoft/playwright | Skill | 99k | yesterday | Apache-2.0 | A | 19 | Claude Code |
| local-ai-agentsmicrosoft/ai-agents-for-beginners | Skill | 79k | 4d ago | MIT | A | 200 | Claude Code, Codex |
| systematic-debuggingobra/superpowers | Skill | 313k | 3d ago | MIT | A | 21 | Claude Code |
| next-cache-components-adoptionvercel/next.js | Skill | 143k | today | MIT | A | 95 | Claude Code |
| next-partial-prefetching-adoptionvercel/next.js | Skill | 143k | today | MIT | A | 103 | Claude Code |