propagation-trace

Trace the origin and spread of news, rumors, and claims using web search.

731|135|Updated Feb 7, 2026
One-click install
npx skills add https://github.com/microclaw/microclaw --skill propagation-trace
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: propagation-trace
Source: https://github.com/microclaw/microclaw/tree/main/skills/built-in/propagation-trace
Command: npx skills add https://github.com/microclaw/microclaw --skill propagation-trace

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

When a news story, rumor, meme, or claim goes viral, it is hard to know where it actually started, who reported it first, and how the wording changed as it spread. This Skill reconstructs the full propagation chain so you can separate the original source from later amplifiers.

Core Features & Use Cases

  • Origin Identification: Walks backward through citations and date-narrowed searches to find the earliest source that is not citing someone else.
  • Chronological Chain Building: Produces a timestamped timeline from origin through early amplifiers to mainstream pickup, with URLs and what each node added.
  • Mutation and Debunk Detection: Compares wording at the origin versus later versions, flagging exaggerations, corrections, and debunks along the way.
  • Use Case: A viral claim about a product recall is circulating on social media. Use this Skill to find the first report, verify its timestamp, map which outlets amplified it, and assess whether the claim mutated or was debunked.

Quick Start

Trace where the claim about the data breach at Company X first appeared and how it spread across news outlets.

Frequently Asked Questions about propagation-trace

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I find the original source of a news story or rumor?

Search for the most distinctive exact phrase or quote from the claim, then re-search with date-narrowing terms and qualifiers like "first reported" or the year. Fetch each candidate page and follow its citations upstream until you reach a source that is not citing anyone else.

How to trace how a claim spread across websites?

Start from the identified origin, then map forward which outlets, aggregators, and platforms amplified it and in what order. Order every node by its verified timestamp and note where the wording changed, was exaggerated, or was corrected.

Does propagation tracing require an API key or special tools?

No API key is required. The method uses only the built-in web_search and web_fetch tools, works on all platforms, and needs no external CLI or paid service.

Why can't I find the true origin of a viral claim?

Web search cannot see private, deleted, or paywalled posts, so the true origin may be invisible. When backward searches repeatedly converge on the same earliest accessible source, treat it as the origin and state the remaining uncertainty explicitly.

What are the limitations of tracing information spread with web search?

Search snippets alone are unreliable, so every node must be verified by fetching the actual page for its timestamp and wording. Appearing earlier does not prove a source caused the spread, and gaps in the chain should be reported rather than filled with guesses.