verifying-citations

verifying-citations is a skill for Claude Code from mronkko/claude-academic-research. It costs 0 tokens per session (2,186 once invoked), scanned A, original, MIT.

A procedure for checking whether an existing citation honestly supports the claim attached to it. It starts with the paper’s abstract and reads the full paper when the claim needs more evidence.

In plain words
What is it for?
Auditing citations in manuscripts, including quoted material, specific statistical results, and research-method claims.
Why use it?
It helps catch citations that are only loosely related to a statement or that do not support its specific details. It keeps citation checking separate from adding new citations.

Skill for Claude Code

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

Part of the academic-research plugin — 11 skills, 1 plugin shipped together

Good fit Auditing citations in manuscripts, including quoted material, specific statistical results, and research-method claims.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/mronkko/claude-academic-research/verifying-citations
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 mronkko/claude-academic-research --skill verifying-citations
Clone the repo
git clone --depth 1 https://github.com/mronkko/claude-academic-research

Made for: Claude Code.

Or install academic-research, the plugin that ships this one along with the rest of its 11 skills, 1 plugin.

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 verifying-citations

README.md
[![agentmods](https://agentmods.dev/badge/skills/mronkko/claude-academic-research/verifying-citations/github.svg)](https://agentmods.dev/skills/mronkko/claude-academic-research/verifying-citations)
Your own site
<a href="https://agentmods.dev/skills/mronkko/claude-academic-research/verifying-citations"><img src="https://agentmods.dev/badge/skills/mronkko/claude-academic-research/verifying-citations/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 verifying-citations

Your own site · 80×15
<a href="https://agentmods.dev/skills/mronkko/claude-academic-research/verifying-citations"><img src="https://agentmods.dev/badge/skills/mronkko/claude-academic-research/verifying-citations.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 0 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 2,186 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. Third-party audits
  • NVIDIA SkillSpector warn 7 Sept 2026
SkillSpector: 1 finding, up to high

These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →

  • high Rogue Agent · line 104
    Skill modifies its own code, configuration, or behavior at runtime. Self-modification enables an agent to escalate privileges, disable safety constraints, or install persistent backdoors.
    Fix: Prevent the skill from modifying its own code, SKILL.md, or configuration files. Treat skill files as read-only at runtime.
How audits are shown
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 $0.00000 $0.02186
Opus 5 $0.00000 $0.01093
Sonnet 5 $0.00000 $0.00437
Haiku 4.5 $0.00000 $0.00219

Measured 10d ago against content hash c7fe6513d3d7, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-10, from the pricing page.

Security

Grade A, and why

verifying-citations 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 10d 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.

skills/verifying-citations/SKILL.md · 144 lines

How it starts

The opening of the file, as written. The whole thing — 144 lines — stays where its author put it; the contents beside it link to each section on GitHub.

verifying-citations

Overview

A citation is verified when an externally consulted source (Zotero PDF fulltext or abstract — never training memory) directly supports the claim the manuscript attributes to it. Verification is staged: read the abstract first; escalate to full text only when the abstract is silent, ambiguous, or insufficient for the kind of claim being made.

This skill is the single source of truth for citation-verification doctrine. fact-check (one-shot audits) and the evidence critic in critic-loop (iterative revision) both load it so that when the rules change, two skills don't drift.

Insertion vs verification. The complementary skill grounded-citations governs inserting a citation during drafting — same externalised-consultation rule, but it escalates on recency / context staling (a faded context-window memory needs re-fetching). This skill governs auditing an existing citation, and it escalates on claim type — quoted passages, specific β coefficients, and method details always require fulltext regardless of what the abstract says. Same demand for grounding; different decision axis.

Unit of dispatch: one subagent per unique citation, not per mention. A paper cited five times in a manuscript becomes one subagent (with five mentions in its prompt), not five subagents. This (a) fetches the source once and (b) lets the subagent catch cross-mention inconsistency — e.g. the manuscript citing the same paper for X in Methods and ¬X in Discussion — which per-mention dispatch can't see. See Multiple mentions of one source below.

Classification scheme

Classification When to assign
VERIFIED The source directly supports the claim — sign, direction, finding, and magnitude all match.
MINOR The claim overreaches the source: oversimplified finding, missing caveat the source considers important, paraphrase that mildly overstates.
MAJOR The paper doesn't exist under that key; wrong paper for the key; direction reversal; fabricated quote; specific finding not supported by the source. Blocks publication.
UNVERIFIABLE Verification requires the full text and no PDF is attached to the Zotero item. The issue is access, not accuracy — recommend running enrich_pdfs.py to populate the library, or replacing the citation. Callers may map UNVERIFIABLE to MAJOR for their own report when the workflow can't ship an unresolved citation (e.g. inside critic-loop).

Read the full file on GitHub · 144 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. 10d ago First seen · 144 lines · 0 tokens per session scan A c7fe6513d3d7

Subscribe to this mod's changes

verifying-citations is a skill published in the GitHub repository mronkko/claude-academic-research (23 stars, last pushed 6d ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 2,186 tokens. 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.

Related

Other skills, from other repositories

systematic-debugging

Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.

obra/superpowers · 21 tokens

local-ai-agents

Build local-first AI agents that run entirely on a developer workstation with Microsoft Foundry Local and Qwen function-calling models. Covers Small Language Models (SLMs), the OpenAI-compatible local endpoint, sandboxed local tools, local RAG with Chroma, local MCP servers, hybrid cloud/local routing, and the…

microsoft/ai-agents-for-beginners · 200 tokens

next-cache-components-adoption

Turn on Cache Components in a Next.js app and resolve the blocking routes it surfaces. Use when the user wants to enable, adopt, or migrate to Cache Components, flip the cacheComponents flag, work through a flood of blocking-prerender / instant validation errors, run the cache-components-instant-false codemod, or…

vercel/next.js · 95 tokens

insight-error-page

Write or audit an insight-kind error page for the Next.js dev overlay. Use when creating a new errors/ .mdx page, auditing an existing one, or checking that a page matches the framework fix cards. Covers page structure, title alignment, FixCard cards with Copy prompt button, code snippets, terminology verification…

vercel/next.js · 83 tokens

next-cache-components-optimizer

Drive a Next.js route to instant navigation by setting up an agentic loop, under Cache Components / PPR, on initial load (hard navigation) and client-side navigation (soft navigation). Encode the goal as a failing @next/playwright instant() e2e and work it to green, one verified route at a time; the shipped test then…

vercel/next.js · 170 tokens

next-partial-prefetching-adoption

Turn on Partial Prefetching in a Next.js app and work through the insights it surfaces. Use when the user wants to enable or adopt Partial Prefetching, flip the partialPrefetching flag, opt routes in with export const prefetch = 'partial', audit Link prefetch={true} behavior, preserve existing prefetched UI with…

vercel/next.js · 103 tokens