Borrowing it
Nothing to install: this file belongs to dengls24/annota. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/dengls24/annota/master/.claude/skills/annota-annotate/SKILL.mdgit clone --depth 1 https://github.com/dengls24/annotaWrote 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/dengls24/annota/annota-annotate)<a href="https://agentmods.dev/skills/dengls24/annota/annota-annotate"><img src="https://agentmods.dev/badge/skills/dengls24/annota/annota-annotate.svg" alt="Measured on agentmods" 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 | $0.00106 | $0.01540 |
| Opus 5 | $0.00053 | $0.00770 |
| Sonnet 5 | $0.00021 | $0.00308 |
| Haiku 4.5 | $0.00011 | $0.00154 |
Grade A, and why
annota-annotate scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 8d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 146 lines — stays where its author put it; the contents beside it link to each section on GitHub.
论文智能标注
你将对一篇学术论文 PDF 执行智能标注。严格遵循两阶段工作流。
输入解析
用户输入:$ARGUMENTS
从输入中提取:
- 目标 PDF:文件路径、论文标题、或 item key
- 标注内容(可选):要标注什么(如"发现结果"、"方法"、"limitation"等)
- 标注颜色(可选):如果用户指定了颜色则使用,否则按语义自动选择
如果用户没有指定标注内容,默认执行"四色全覆盖标注"(见下方默认模式)。
颜色语义映射
| 内容类型 | 颜色 | 代码 |
|---|---|---|
| 核心贡献、创新点、novelty | 紫色 | #a28ae5 |
| 方法、算法、模型、架构设计 | 蓝色 | #2EA8E5 |
| 实验结果、发现、数据、性能 | 绿色 | #28CA42 |
| 问题、不足、limitation、future work | 红色 | #ff6666 |
| 通用/用户未指定单一类型 | 黄色 | #ffd400 |
Phase 1 — 理解论文内容
Step 1.1:定位 PDF
如果用户提供了文件路径,直接使用。否则:
- 调用
search_zotero_items(query)搜索论文 - 获取
pdf_attachment_id
Step 1.2:提取全文
调用 get_pdf_text_bulk(item_id, skip_refs=True) 批量提取纯文本。
Step 1.3:分析内容
如果用户指定了标注目标,只识别该类内容。
如果用户没有指定(默认四色全覆盖模式),识别以下四类,每类至少找 3-5 处最有代表性的句子:
| 类型 | 标准 | 最少数量 |
|---|---|---|
| 贡献/创新点 | 论文声称的 novelty、"we propose"、"our key insight" | 3-5 处 |
| 方法/架构 | 核心技术设计、关键机制、算法描述 | 3-5 处 |
| 实验结果 | 具体数字、性能对比、improvement 数据 | 4-6 处 |
| 局限/不足 | limitation、weakness、future work | 2-3 处 |
输出标注清单(在对话中展示):
📋 将创建以下标注(共 N 条):
[紫·贡献] Page 1: "We propose DCC, a novel..."
[紫·贡献] Page 2: "Our key insight is..."
[蓝·方法] Page 3: "The architecture consists of..."
[蓝·方法] Page 4: "We design a two-stage pipeline..."
[绿·结果] Page 6: "achieves 2.3× speedup over baseline"
[绿·结果] Page 6: "reduces energy by 41%"
[红·不足] Page 8: "Our approach is limited to..."
...
⚠️ 写入需要关闭 Zotero。请确认已关闭后回复"确认",或直接说"开始"。
Step 1.4:检查已有标注
调用 list_annotations(item_id) 查看是否已有标注,避免重复。
如果已有大量标注,告知用户并询问是否继续(追加)。
Phase 2 — 精确标注
Step 2.1:获取目标页坐标
只对包含目标句子的页面调用 get_pdf_layout_text(item_id, page_number)。
绝对不要对每一页都调用这个工具——只调用目标页。 多个目标句子在同一页时,只调用一次该页。
Step 2.2:匹配坐标
在返回的 blocks 中找到目标句子对应的 rect 坐标。 一个句子可能跨多行,需要收集所有匹配行的 rect。 匹配时用子串匹配(句子的前 30 个字符即可定位),不要精确匹配。
Step 2.3:写入标注
用户确认后,调用 batch_annotate 一次性写入所有标注:
{
"item_id": "...",
"annotations": [
{"page_index": 0, "rects": [...], "color": "#a28ae5", "text": "贡献描述..."},
{"page_index": 2, "rects": [...], "color": "#2EA8E5", "text": "方法描述..."},
{"page_index": 5, "rects": [...], "color": "#28CA42", "text": "结果数据..."},
{"page_index": 7, "rects": [...], "color": "#ff6666", "text": "局限说明..."}
]
}
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.
- 8d ago First seen · 146 lines · 106 tokens per session scan A d4d9f5862f2d
annota-annotate is a skill published in the GitHub repository dengls24/annota (19 stars, last pushed 4mo ago), licensed MIT. It adds 106 tokens to every session and 1,540 once invoked, about $0.0005 per session on Opus 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.
Other skills, from other repositories
latex-compile
Compile a LaTeX document and fix every error plus aesthetic issue (overfull/underfull boxes, widows, alignment, fonts) for a clean PDF and log. Use this instead of running pdflatex/latexmk manually — it avoids the latexmk stale-log trap and silent grep failures on binary log output, and it reformats rather than…
latex-document-skill
Universal LaTeX document skill: create, compile, and convert any document to professional PDF with PNG previews. Supports resumes, reports, cover letters, invoices, academic papers, theses/dissertations, academic CVs, presentations (Beamer), scientific posters, formal letters, exams/quizzes, books, cheat sheets…
phd-deepread
Process academic PDFs into structured Obsidian literature notes and 9-node critical-thinking canvases. Activate when the user shares a PDF (drag-in or path) and asks to read, summarize, analyze, or "deeply read" it — or when they say things like "phd-deepread read this paper", "make a literature note", "critique this…
PDF Content Extractor & Annotator
A tool for extracting and analyzing PDF content, including text, tables, metadata, and embedded images. It can also combine PDFs, add annotations, remove some watermarks, and use OCR to read scanned pages.
clinical-reports
Write comprehensive clinical reports including case reports (CARE guidelines), diagnostic reports (radiology/pathology/lab), clinical trial reports (ICH-E3, SAE, CSR), and patient documentation (SOAP, H&P, discharge summaries). Full support with templates, regulatory compliance (HIPAA, FDA, ICH-GCP), and validation…
Use this skill whenever the user wants to do anything with PDF files. This includes reading or extracting text/tables from PDFs, combining or merging multiple PDFs into one, splitting PDFs apart, rotating pages, adding watermarks, creating new PDFs, filling PDF forms, encrypting/decrypting PDFs, extracting images, and…