cross-source-verify

cross-source-verify is a cursor rule for coding agents from ilyautov/small-business-ru. It costs 21 tokens per session (522 once invoked), scanned A, original, Apache-2.0.

A rule for checking facts against multiple sources and clearly showing when they disagree. It ranks sources by authority and freshness and separates confirmed information from uncertain signals.

In plain words
What is it for?
Verifying company data, comparing dates and values across sources, deduplicating records, and reporting agreement or disagreement with source quality and confidence.
Why use it?
It helps prevent confident but incorrect conclusions based on one source or on conflicting figures.

Cursor rule

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.

agentmods
npx agentmods add rules/ilyautov/small-business-ru/cross-source-verify
Clone the repo
git clone --depth 1 https://github.com/ilyautov/small-business-ru

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 cross-source-verify

README.md
[![agentmods](https://agentmods.dev/badge/rules/ilyautov/small-business-ru/cross-source-verify.svg)](https://agentmods.dev/rules/ilyautov/small-business-ru/cross-source-verify)
Your own site
<a href="https://agentmods.dev/rules/ilyautov/small-business-ru/cross-source-verify"><img src="https://agentmods.dev/badge/rules/ilyautov/small-business-ru/cross-source-verify.svg" alt="Measured on agentmods" height="20"></a>
Per session 21 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 522 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
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 $0.00021 $0.00522
Opus 5 $0.00010 $0.00261
Sonnet 5 $0.00004 $0.00104
Haiku 4.5 $0.00002 $0.00052

Measured 4d ago against content hash 8b7aeb80cb33, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

cross-source-verify 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 4d 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.

adapters/cursor/cross-source-verify.mdc · 30 lines

What it actually says

Cross-Source Verify — круговая сверка (Cursor)

Скопируй в .cursor/rules/. Последняя миля любой многоисточниковой проверки.

ПРИНЦИП: один источник звучит уверенно и часто врёт; конкретная цифра ≠ верная. В живом тесте 5 агрегаторов по одной компании дали число судов почти втрое разное (~500 / ~1000 / ~1500), при этом выручка и ФССП совпали у всех. Факт = совпадение у ≥3; расхождение = разброс, а не выбор одной версии. Цифры округлены.

УВЕРЕННОСТЬ = свежесть × авторитет × согласие.

  • Авторитет (РФ): госреестр (ЕГРЮЛ/ГИР БО/kad.arbitr/ФССП/ЕФРСБ) > офиц. сводка > агрегатор проверенный (checko/rusprofile/saby/audit-it) > агрегатор шумный > форум/отзыв > «со слов».
  • Свежесть: у каждого факта дата; устаревшее понижай.
  • Согласие: совпало у ≥3 → ✅ прямое утверждение; разошлось → ⚠️ покажи разброс; сигнал в одном источнике (особенно тревожный) → проверь принадлежность (ИНН/ОГРН), не пугай.

ДЕДУП по приоритету полнота > авторитет > свежесть. Не сливай разные выводы/периоды.

ВЫВОД: связный нарратив по фактам, не «у источника A… у B…». Конфликт показывай явно (значения, даты, tier, вероятная причина, что надёжнее). Каждый факт → источник + дата + tier (✅/⚠️/❌). Никаких чисел без происхождения. Конфликт не прячь, атрибуцию не опускай.

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. 4d ago First seen · 30 lines · 21 tokens per session scan A 8b7aeb80cb33

Subscribe to this mod's changes

cross-source-verify is a cursor rule published in the GitHub repository ilyautov/small-business-ru (15 stars, last pushed 2mo ago), licensed Apache-2.0. It adds 21 tokens to every session and 522 once invoked, about $0.0001 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 2026-08-30.

Related

Other cursor rules, from other repositories

dependencies

Dependency management guidelines.

raw-labs/mxcp · 86 tokens

remote-pc

Drive a host running agentic-remote-pc over its MCP runner. Use when the user asks to run commands on a remote PC, drive a coding-agent CLI on a host, or validate a local feature/deployment.

cogniziocompany/agentic-remote-pc · 0 tokens

multi-source-signal-synthesiser

Synthesises user signals from multiple research sources into a unified, weighted insight brief. Use when you have data from interviews, support tickets, NPS verbatims, app reviews, or sales calls and need to reconcile contradictions, surface the underlying need behind requests, or answer 'what are users really telling…

mohitagw15856/pm-claude-skills · 89 tokens

voice-agent-design

Design a voice AI agent for phone or in-app conversations — call flows, interruption handling, escalation to humans, and the metrics that catch a bad voice experience. Use when asked to design a voice agent, automate a phone line, spec an IVR replacement, or review why callers hate an existing voice bot. Produces a…

mohitagw15856/pm-claude-skills · 95 tokens

agent-spec

Specify an autonomous or tool-using AI agent before building it. Use when asked to design an AI agent, define an agent's tools and guardrails, scope what an agent is allowed to do, or write an agent spec/PRD. Produces an agent spec — goal & scope, tools with permissions, the control loop, guardrails & approval gates…

mohitagw15856/pm-claude-skills · 85 tokens

ai-content-audit

Audit a content library, docs site, or blog for AI-generated filler that's eroding trust and search performance — and triage what to fix, rewrite, or delete. Use when asked to find slop in a content library, audit AI-written content quality, explain why content engagement or rankings dropped after scaling with AI, or…

mohitagw15856/pm-claude-skills · 132 tokens