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 agentmods add skills/redhuntlabs/wizard/verifying-before-citingnpx skills add redhuntlabs/wizard --skill verifying-before-citinggit clone --depth 1 https://github.com/redhuntlabs/wizardWrote 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/redhuntlabs/wizard/verifying-before-citing)<a href="https://agentmods.dev/skills/redhuntlabs/wizard/verifying-before-citing"><img src="https://agentmods.dev/badge/skills/redhuntlabs/wizard/verifying-before-citing.svg" alt="Measured on agentmods" 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 | $0.00031 | $0.01127 |
| Opus 5 | $0.00015 | $0.00563 |
| Sonnet 5 | $0.00006 | $0.00225 |
| Haiku 4.5 | $0.00003 | $0.00113 |
Grade A, and why
verifying-before-citing 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 5d 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 — 107 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Verifying Before Citing
What this does
Forces every cited claim, statistic, quote, or attribution to be verified against the actual primary source before it appears in writing. Eliminates "secondhand citation drift" — the failure mode where a claim is true at the start of a citation chain but mangled by the third repetition.
When to use
- About to publish, ship, or send anything that cites a source
- About to attribute a quote to a person
- About to cite a statistic, percentage, or numeric finding
- About to summarize what a source "says" without having opened it recently
- Whenever pulling from memory, AI output, or a secondary source
What you bring (Inputs)
- The claim and its proposed citation
- Access (or lack of access) to the primary source
What you get (Output)
A claim that you have personally verified against the primary source, OR an explicit hedge ("according to X, who cites Y, which I have not verified") OR the claim removed.
How it works (Steps)
This is a discipline. The behavioral structure is: a non-negotiable rule (below), an excuses-and-counters table to defeat rationalizations, warning signs to spot the moment of risk, and a hard gate that fires before publishing. Apply the gate at every citation; verify, hedge, or cut.
The non-negotiable rule
Every cited claim, quote, statistic, or attribution must be verified against the primary source — opened, read in context — before it leaves your hands. If you cannot, you must hedge explicitly or remove the claim.
Excuses and counters
| If you think... | Reality |
|---|---|
| "I remember reading this" | Memory drifts. Open the source. |
| "I've cited this before" | The source may have updated, or your prior cite may have been wrong. Open the source. |
| "An AI told me this is from X" | AI-attributed quotes are routinely fabricated or misattributed. Verify or remove. |
| "It's a famous quote, everyone knows it's from X" | Famous quotes are misattributed at the highest rates. Especially these. Verify. |
| "I'm short on time" | The cost of a wrong citation lasts longer than the time to verify it. Verify. |
| "It's a small detail" | Small wrong details destroy credibility for the rest. Verify. |
| "The secondary source seems reliable" | Even reliable secondaries get details wrong. Open the primary. |
| "I'll mark it for fact-check later" | "Later" doesn't happen. Verify now or hedge now or cut now. |
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.
- 5d ago First seen · 107 lines · 31 tokens per session scan A 474ce7acb925
verifying-before-citing is a skill published in the GitHub repository redhuntlabs/wizard (9 stars, last pushed 4mo ago), licensed MIT. It adds 31 tokens to every session and 1,127 once invoked, about $0.0002 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
fastapi
FastAPI - 高性能 Python Web API 框架.
fzf
Skill "fzf" from CN-big-cabbage/github-skill-distiller, covering fzf 通用命令行模糊查找器, 技能概述, 使用流程, 关键章节导航 and ai 助手能力.
glances
Skill "glances" from CN-big-cabbage/github-skill-distiller, covering glances 跨平台系统监控工具, 技能概述, 使用流程, 关键章节导航 and ai 助手能力.
lazydocker
本技能帮助用户通过 lazydocker 的终端 UI 界面管理 Docker 容器、镜像、卷和网络，支持以下场景：.
paper-checking
论文查重系统 - 一亿字次级论文库秒级查重，支持纵向查重和横向查重.
you-get
网页媒体下载助手 - 从YouTube、Bilibili等网站下载视频、音频、图片.