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 oscardvs/zoteus --skill evidence-tablegit clone --depth 1 https://github.com/oscardvs/zoteusWrote 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/oscardvs/zoteus/evidence-table)<a href="https://agentmods.dev/skills/oscardvs/zoteus/evidence-table"><img src="https://agentmods.dev/badge/skills/oscardvs/zoteus/evidence-table/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/oscardvs/zoteus/evidence-table"><img src="https://agentmods.dev/badge/skills/oscardvs/zoteus/evidence-table.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 | $0.00074 | $0.00942 |
| Opus 5.5 | $0.00030 | $0.00377 |
| Sonnet 5.5 | $0.00015 | $0.00188 |
| Haiku 4.5 | $0.00007 | $0.00094 |
Grade A, and why
evidence-table 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 19d 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 — 36 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Compare sources, review a topic, or audit citations from the Zotero library
These steps use the Zoteus tools, named zotero_*. The rule that holds throughout: every quotation and page comes from a zotero_get_fulltext call made in this conversation. A table built from search snippets looks the same as a real one and cannot be cited.
Evidence table
- Choose the studies. Default to at most 5 unless the user asks for more; each costs one full-text call.
- Given item keys: call
zotero_get_itemon each for the title, creators and date, and search for nothing else. - Given a collection: resolve its name to a key with
zotero_list_collections, then callzotero_search_itemswithcollectionKeyandtop:trueso child notes and attachments do not become rows.zotero_semantic_searchcannot be scoped to a collection. - Otherwise:
zotero_semantic_searchwithqset to the question, falling back tozotero_search_itemswithtop:truewhen meaning search is unavailable.
- Given item keys: call
- Retrieve one passage per study. Call
zotero_get_fulltextwith the item key andqueryset to the question, one item at a time. Take the quotation verbatim frompassages[0].text. Usepagewhen it is present andpageApproxotherwise, and keep that distinction in the table. When neither is present, leave the locator empty and say so. - Classify coverage from what the tools returned. Use
passagewhen a passage came back,abstract-onlywhen onlyzotero_get_itemreturned anabstractNote, andunavailablewhen neither did. - Judge support. The finding and the support rating are your judgement, so word them as judgement. Rate support as
supported,contradicted,uncertainorunverified. A row with no retrieved passage isunverified, never blank and neversupported. - Render with the tool. Pass the question and one row per study to
zotero_evidence_tablewithformatset to"markdown"or"csv". Do not type the table out yourself: the tool counts the coverage summary from the rows and names any row that contradicts its own evidence. Repeat its warnings as given. Fix the row or leave the gap visible, and never edit a quotation to silence a warning. Passsave_pathonly if the user asked for a file. - Add the references for the same item keys with
zotero_bibliography. - Ask before saving anything into the library.
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.
- 19d ago First seen · 36 lines · 74 tokens per session scan A 7f2c5dcb9bd4
evidence-table is a skill published in the GitHub repository oscardvs/zoteus (58 stars, last pushed 10d ago), licensed MIT. It adds 74 tokens to every session and 942 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.
Other skills, from other repositories
get-research-paper
Discovers, retrieves, ranks, and summarizes real existing research papers on any topic. Searches arXiv, Google Scholar, PubMed, Semantic Scholar, and reputable open repositories; returns a curated reading list with verified DOIs, key findings, and citation-ready metadata. Activates on slash commands…
science-agent
Verify what AI writes about science — catch confabulated citations, drifted notebook numbers, and stale figure captions before they ship. Use when adding or reviewing academic citations, BibTeX entries, DOIs, or empirical claims in .md/.tex/.bib/.ipynb files; when writing a literature review; or when the user says…
drug-discovery
Drug discovery: ChEMBL search, drug-likeness, interactions.
research-engineer
An uncompromising Academic Research Engineer. Operates with absolute scientific rigor, objective criticism, and zero flair. Focuses on theoretical correctness, formal verification, and optimal implementation across any required technology.
build-interactive-explainers
A guide for building interactive explainers, calculators, and simulations driven by an executable model. Users change inputs, steps, states, or events to understand a rule or see how a process develops over time.
arxiv
Search arXiv papers by keyword, author, category, or ID.