investigate-anything

Frames OSINT investigations with scope, collection plans, source grading, and hypothesis testing.

38|2|Updated Aug 2, 2026
One-click install
npx skills add https://github.com/UseOSINT/Skills --skill investigate-anything-useosint
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: investigate-anything
Source: https://github.com/UseOSINT/Skills/tree/main/skills/investigate-anything
Command: npx skills add https://github.com/UseOSINT/Skills --skill investigate-anything-useosint

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve? Open-source investigations often fail at the framing stage: vague objectives, unauthorized scope, circular corroboration, and confirmation bias produce confident but wrong findings. This Skill turns a vague request into an answerable intelligence question and enforces disciplined tradecraft before any collection begins. ## Core Features & Use Cases - Scope and Objective Framing: Defines authorized boundaries, rewrites vague requests into falsifiable questions, and writes a passive-first collection plan mapped to sources. - Workflow Routing: Routes any starting selector (name, company, domain, handle, email, phone, photo, social profile) to the correct investigation workflow skill. - Source Grading and Hypothesis Testing: Applies the Admiralty (NATO) source-grading scheme and Analysis of Competing Hypotheses, with reference worksheets and worked examples. - Use Case: Given a supplier invoice, frame the question "does this supplier have an undisclosed sanctioned owner", plan passive collection, route to a company investigation workflow, grade each finding, and test competing attribution hypotheses before reporting. ## Quick Start Ask the agent to investigate a person, company, or domain and it will set authorized scope, frame an answerable question, and route to the right workflow.

Frequently Asked Questions about investigate-anything

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

FAQPage Schema
How do I start an OSINT investigation on a person or company?▼

Start by defining authorized scope: subject, objective, in-bounds and out-of-bounds sources, and jurisdiction. Then rewrite the request into a falsifiable question, write a passive-first collection plan, and route your starting selector to the matching workflow skill.

What is the Admiralty source grading system in OSINT?▼

The Admiralty (NATO) scheme grades each item with a letter for source reliability (A-F) and a number for information credibility (1-6), such as B2. The two axes are graded independently so a reliable source passing unverified information stays distinct from confirmed facts.

How do I avoid confirmation bias in open-source investigations?▼

Use Analysis of Competing Hypotheses: list all hypotheses including boring ones, mark evidence as consistent or inconsistent per hypothesis, and pick the hypothesis with fewest inconsistencies. Evidence consistent with every hypothesis has no diagnostic value and should be discarded.

When is it legal to investigate someone using public data?▼

Public availability is not permission. In the EU and UK, aggregating public facts about a living person is processing personal data and requires a lawful basis. Never authenticate to, probe, or contact systems belonging to the subject without written authorization.

Why do multiple agreeing sources not confirm an OSINT finding?▼

Multiple sites often copy one origin or resell the same data feed, which is circular reporting rather than corroboration. Check publication order, shared phrasing or typos, and upstream data feeds before grading anything as confirmed.