grounded-answers

grounded-answers is a skill for Claude Code from oscardvs/zoteus. It costs 70 tokens per session (788 once invoked), scanned A, original, MIT.

A guide for answering questions from papers saved in Zotero, a reference manager for research sources. It finds the relevant item and supplies verbatim passages, page numbers, figures, or tables from its PDF.

In plain words
What is it for?
Use it to check what a saved paper says, locate the source of a claim, review highlights and notes, or retrieve a passage, figure, or table.
Why use it?
It grounds answers in the user's own papers rather than unsupported summaries or memory. It also explains what to do when the Zotero connection or library is unavailable.

Skill for Claude Code

Written for Claude Code: shipped in a Claude Code plugin. Also seen: mentions Claude Code.

Part of the zoteus plugin — 4 skills, 1 MCP server shipped together

Good fit Use it to check what a saved paper says, locate the source of a claim, review highlights and notes, or retrieve a passage, figure, or table.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/oscardvs/zoteus/grounded-answers
Install

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.

Any agent
npx skills add oscardvs/zoteus --skill grounded-answers
Clone the repo
git clone --depth 1 https://github.com/oscardvs/zoteus

Made for: Claude Code.

Or install zoteus, the plugin that ships this one along with the rest of its 4 skills, 1 MCP server.

Wrote 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.

agentmods badge for grounded-answers

README.md
[![agentmods](https://agentmods.dev/badge/skills/oscardvs/zoteus/grounded-answers/github.svg)](https://agentmods.dev/skills/oscardvs/zoteus/grounded-answers)
Your own site
<a href="https://agentmods.dev/skills/oscardvs/zoteus/grounded-answers"><img src="https://agentmods.dev/badge/skills/oscardvs/zoteus/grounded-answers/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.

agentmods 80×15 button for grounded-answers

Your own site · 80×15
<a href="https://agentmods.dev/skills/oscardvs/zoteus/grounded-answers"><img src="https://agentmods.dev/badge/skills/oscardvs/zoteus/grounded-answers.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 70 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 788 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5.1 $0.00070 $0.00788
Opus 5.5 $0.00028 $0.00315
Sonnet 5.5 $0.00014 $0.00158
Haiku 4.5 $0.00007 $0.00079

Measured 25d ago against content hash 0480ea81102f, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

grounded-answers 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 25d 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.

plugins/zoteus/skills/grounded-answers/SKILL.md · 25 lines

How it starts

The opening of the file, as written. The whole thing — 25 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Answer from the Zotero library, with the passage to show for it

These steps use the Zoteus tools, named zotero_*. If none are available, tell the user the Zoteus server is not running. In Claude Code, and in Cowork on their computer, this plugin starts it, which needs Node.js 20.19 or later and either the Zotero desktop app running or a Zotero API key in the plugin's settings. In claude.ai chat, a local server cannot run, so they need a remote Zoteus connector instead.

Steps

  1. Confirm the library. On the first library request in a conversation, call zotero_whoami. It reports which library calls default to and whether the desktop app or the cloud API serves it. If the user means a group library, get its numeric id from zotero_groups and pass library_type:"group" and library_id on every later call. Never ask the user to type a numeric user id.
  2. Find the item. For a known title, author or year, use zotero_search_items with q, which is instant and needs no index. For a topic or a question, use zotero_semantic_search with q. When its index is empty it starts a background build and says so: tell the user, and meanwhile use zotero_search_items with qmode:"everything", which also searches notes and indexed PDF text. Show the title, first author, year and item key so the user can confirm it is the right item.
  3. Retrieve the passage. Call zotero_get_fulltext with the item key and query set to the claim or question. Quote passages[].text verbatim. Give the page and say what kind it is: page is exact, pageApprox is a proportional estimate, and pageSource says which one this call produced. For a long document, call it with outline:true first and then with page_range for the pages the outline names.
  4. Look at what text extraction loses. For a figure, a table, an equation or a scanned page, call zotero_pdf_images with mode:"pages" and pages set to that page, and read the image.
  5. Report coverage. If the full text could not be retrieved, say so and say what you worked from instead: the abstract from zotero_get_item, a note, or nothing.

Read the full file on GitHub · 25 lines

Changes

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.

  1. 25d ago First seen · 25 lines · 70 tokens per session scan A 0480ea81102f

Subscribe to this mod's changes

grounded-answers is a skill published in the GitHub repository oscardvs/zoteus (61 stars, last pushed 3d ago), licensed MIT. It adds 70 tokens to every session and 788 once invoked, about $0.0003 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 2026-09-28.

Related

Other skills, from other repositories

pdf

PDF files: create, read, merge, fill, OCR, edit text.

NousResearch/hermes-agent · 18 tokens

wowerpoint

Turn one document into a kawaii NotebookLM slide deck. Use for "wowerpoint this", "make a deck about ", "turn this report into slides", or any request to render a single document as shareable narrative slides.

thedotmack/claude-mem · 53 tokens

save-as-pdf

Reformat the current HTML design into a paginated, paper-ready PDF. The "Instant" export already gives the user a PDF at the design's native pixel size — this path is for when they want real pages.

asgeirtj/system_prompts_leaks · 9 tokens

pdf

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…

asgeirtj/system_prompts_leaks · 92 tokens

docs-artifact

Use when creating, editing, or reviewing a document for the user (Word, Google Docs, or a document delivered as PDF), or when the task calls for a standalone written deliverable such as a memo, report, proposal, or brief. Also use to answer questions that require inspecting an existing document; not for a short chat…

asgeirtj/system_prompts_leaks · 84 tokens

form-filling

Fill out online forms (Google Forms, Typeform and similar) or PDF, Word and DocuSign forms; prepare a reviewable draft and get approval before submitting or sending.

asgeirtj/system_prompts_leaks · 40 tokens