· 1 observation
Technical updates · 1
Package or technical metadata updated
View technical comparison →Delores v1.5.0
AI-guided close paper reading
From the developer
Not specified in the available declarations: limitations, pricing & account requirements.
Read this paper section by section with cited source locations.
Developer declarations; not independently tested. Listing last checked Nov 9, 2026. Package observed Nov 9, 2026.
From the marketplace listing
Guide rigorous close reading of academic papers, methods, experiments, figures, tables, and appendices. Explanations stay grounded in source locations and can be turned into reusable research notes. Open source on GitHub: https://github.com/Delores-Lin/paper-close-reading If you find it useful, consider leaving a star. For bugs, questions, or feature requests, please open an issue: https://github.com/Delores-Lin/paper-close-reading/issues
Language: English · Automatically detected from descriptions.
Exact text from the indicated source. A mention alone does not establish support for your task.
Source-grounded close reading of academic papers with traceable page, section, figure, and table references.
· 1 observation
Package or technical metadata updated
View technical comparison →· 2 observations
Package or technical metadata updated
View technical comparison →· 2 observations
Package or technical metadata updated
View technical comparison →Package or technical metadata updated
View technical comparison →Find competitors to follow as you develop your plugin.
Find academic papers and listen to them narrated, section by section, with the text alongside.
Shared terms: paper, academic, figure, section
Research scientific literature, critically read and compare papers, explore evidence gaps, draft cited academic documents, and organize r...
Shared terms: paper, academic, guided, reference
Zoteus connects your Zotero library to ChatGPT. Search your references and the full text of your PDFs, pull passages with the page number...
Shared terms: paper, figure, reference, stay
Find, compare, and synthesize academic research with transparent evidence and citations.
Shared terms: paper, academic, research, question
alphaXiv helps users discover academic papers, read and query paper content, inspect linked GitHub repositories, and manage folders of sa...
Shared terms: paper, academic, github, research
--- name: paper-close-reading description: Look up paper metadata and citation counts with the bounded Google Scholar helper or host-provided browser when requested, and explore citing papers only on request. Guide rigorous, source-grounded close reading of academic papers, technical reports, PDFs, sections, figures, tables, and appendices through either guided interactive reading or autonomous full-paper analysis. Use when the user asks to read a paper closely, continue to the next section, explain selected text, follow a three-pass reading process, inspect cited visual evidence, critique experiments, resume a multi-turn reading, or produce reusable research notes with precise source locations, methods, metrics, limitations, and task transfer. --- # Paper Close Reading ## Google Scholar Citation Workflows - For citation counts (引用量), a requested metadata refresh, or initial paper metadata, read [references/google-scholar-citations.md](references/google-scholar-citations.md). - For the default fast lookup, use [scripts/scholar_lookup.py](scripts/scholar_lookup.py) with its explicit Python 3 standard-library prerequisite. It performs one local Node capability preflight per launcher process, uses Node native `fetch` when available and proxy-compatible, and tries curl only when Node is unavailable, unsupported, or cannot use the configured proxy. A new CLI invocation repeats the local preflight; it never performs a network probe. Do not rotate transports after a 403, 429, CAPTCHA, or other explicit access restriction. - For the structured paper-query path and citation fields, read [references/google-scholar-json.md](references/google-scholar-json.md); use its bounded browser fallback and do not assume structured citing-list support. - For citing papers (被引), year filters, recent additions (最新被引), or impact ranking, read [references/google-scholar-citing.md](references/google-scholar-citing.md). - For an explicitly requested full-text download after a candidate has been matched, use [scripts/scholar_download.py](scripts/scholar_download.py) with the selected candidate's `full_text.url`. A valid PDF file and paper identity/version verification are separate states; never select the first candidate or claim identity from the PDF magic bytes alone. - Both workflows load [references/google-scholar-browser.md](references/google-scholar-browser.md) only when browser interaction is needed. It owns shared timeouts, recovery, and English/Simplified Chinese locators. - A citation-only request does not require reading-mode selection or start the three passes. Preserve reading progress and reuse verified metadata with its source and time; write a persistent cache only when the helper's cache path is explicitly requested. Do not block paper reading on unavailable counts. Match the user's language; English internal instructions do not require English answers. - For Scholar terminal access in a known network-restricted sandbox, use host-authorized sandbox-external execution; after success, reuse that permitted path. For connection failures or suspicious 404 responses, check existing proxy configuration as described in [sandbox and proxy requirements](references/google-scholar-json.md#sandbox-and-proxy-requirements). ## Overview Read papers through a source-grounded three-pass process and produce reusable research outputs. Support guided interactive reading and autonomous full-paper analysis without silently choosing between them. Match the user's language unless the user requests another language. Do not give generic summaries. Rebuild context from the relevant pages before explaining, cite precise locations, interleave short source fragments with explanation in the user's language, integrate cited figures and tables with the surrounding argument, distinguish paper claims from inference or critique, and preserve progress during sustained multi-turn reading. ## Core Workflow 1. Identify the requested scope: one-off explanation, whole-pape
Preview of the first 4,000 characters. Read full instructions →
Referenced files: 11 — file references are metadata, not package contents.
Publisher declarations from the archived package. These are separate from our research and the live service's terms.
Package observed Nov 9, 2026.
plugins_6aaf76e13e4881919760c09b0197f075
Download plugin data (JSON)Plugin details: freshly prepared. Prepared . Refreshed when plugin data changes or after 6 hours. Your account and Follow status are live.