Skill guides · 2 min read
Cross-entity correlation
Acquire relevant public evidence, then correlate evidence-backed relationships.
When to use it
Use this to investigate relationships around a subject across relevant public sources. The Skill can acquire new evidence before correlating entities; earlier evidence in the same case provides context but does not define the entire search scope.
Before you start
Available on Agency. Fill every required field. Example values below illustrate the form; replace them with your intended subject.
| Field | Required | What to enter |
|---|---|---|
| Subject | Required | Enter subject. |
Run the Skill
- In Skill library, find Cross-entity correlation and select Use template.
- Fill in the template fields using your subject’s details, then check the Preview.
- Select Insert into chat to load the prepared request into the investigation composer.
- Check the intended case and send the unchanged request to run this Skill. To change its inputs, return to the template and insert it again.
Subject: Example Research Ltd
Read the results
Inspect each relationship and its supporting evidence. Directly published links, inferred associations, contradictions, and gaps should remain distinguishable. Follow any domain or infrastructure relationship back to the observations that support it. Shared addresses, co-hosting, similar names, or dataset inclusion can be useful leads without establishing common legal ownership, control, or wrongdoing. Keep unresolved entities separate until the evidence supports joining them.