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 heidihelena/citevahti --skill citevahti-screengit clone --depth 1 https://github.com/heidihelena/citevahtiWrote 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/heidihelena/citevahti/citevahti-screen)<a href="https://agentmods.dev/skills/heidihelena/citevahti/citevahti-screen"><img src="https://agentmods.dev/badge/skills/heidihelena/citevahti/citevahti-screen/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/heidihelena/citevahti/citevahti-screen"><img src="https://agentmods.dev/badge/skills/heidihelena/citevahti/citevahti-screen.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 | $0.00075 | $0.01055 |
| Opus 5 | $0.00037 | $0.00528 |
| Sonnet 5 | $0.00015 | $0.00211 |
| Haiku 4.5 | $0.00007 | $0.00105 |
Grade A, and why
citevahti-screen 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 11d 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 — 86 lines — stays where its author put it; the contents beside it link to each section on GitHub.
CiteVahti screen — sweep the reference list before a reviewer does
citevahti-dev checks claims one at a time as you write. This skill sweeps the
references you already have and flags what a reviewer would catch — retractions,
expressions of concern, and citations whose source does not support the sentence. It is a
read-only worklist: nothing is fixed, removed, or written back.
Triggers
Use when the researcher asks to:
- check a manuscript or reference list for retracted papers before submitting
- screen a bibliography for problems in one sweep
- get a triage list of citations that need a second look
Do NOT use when: the ask is to check a single new claim (citevahti-dev), package a
finished audit (citevahti-report), or review someone else's manuscript
(citevahti-review).
What it checks (each a real command; nothing is auto-fixed)
- The citation suite —
citevahti test --online. Runs the "unit tests for citations" over the ledger and, with--online, also checks each cited reference is real and not retracted. This is the single closest thing to "screen the whole manuscript." - Retraction status —
citevahti retraction-scan(bare = scan every staged reference; or--citekey/--doi/--pmidfor one). Cross-checks against retraction and expression-of-concern notices. A clean result is "no known retraction," not a warranty that every one was caught. - Claim–source support —
citevahti claim-check --claim "<sentence>"(deterministic, read-only, against your Zotero library). Flags citations whose source likely does not support the sentence, for the human to check. - Triage worklist —
citevahti triage. Surfaces what needs attention, worst first, each with a plain reason and next action.
There is no single citevahti screen <file> command — screening is this read-only sweep
composed from the commands above, ending in a triage list.
Output — a triage list, ranked by severity
citevahti test --online # citation suite incl. real-and-not-retracted checks
citevahti retraction-scan # retraction / expression-of-concern sweep
citevahti triage # the worst-first worklist to hand back
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.
- 11d ago First seen · 86 lines · 75 tokens per session scan A 7728224e3bd8
citevahti-screen is a skill published in the GitHub repository heidihelena/citevahti (1 stars, last pushed 6d ago), licensed Apache-2.0. It adds 75 tokens to every session and 1,055 once invoked, about $0.0004 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 2026-08-31.
Other skills, from other repositories
ml-paper-writing
Write publication-ready ML/AI papers for NeurIPS, ICML, ICLR, ACL, AAAI, COLM. Use when drafting papers from research repos, structuring arguments, verifying citations, or preparing camera-ready submissions. Includes LaTeX templates, reviewer guidelines, and citation verification workflows.
ml-paper-writing
Write publication-ready ML/AI papers for NeurIPS, ICML, ICLR, ACL, AAAI, COLM. Use when drafting papers from research repos, conducting literature reviews, finding related work, verifying citations, or preparing camera-ready submissions. Includes LaTeX templates, citation verification workflows, and paper…
paper-research
An evidence-gathering guide for social-science research. It finds prior studies, academic theories, official statistics, and relevant cases, while recording sources for claims.
ml-paper-writing
Write publication-ready ML/AI papers for NeurIPS, ICML, ICLR, ACL, AAAI, COLM. Use when drafting papers from research repos, structuring arguments, verifying citations, or preparing camera-ready submissions. For systems venues (OSDI, NSDI, ASPLOS, SOSP), use systems-paper-writing instead.
scientometric-paper-review
A review skill for quantitative social-science papers, especially studies that count publications or estimate effects from panel or policy data. Bibliometrics measures patterns in published research, and a quasi-experiment estimates an effect without a fully controlled experiment.
paper-writing
A writing workflow for producing a social-science research paper from an approved research design, cited prior studies, and analysis results.