investigation_skill

This skill helps you locate, verify, and reconnect with people using public data through parallel, ethical investigations across multiple sources.
  • TypeScript

10.2k

GitHub Stars

1

Bundled Files

3 weeks ago

Catalog Refreshed

1 month ago

First Indexed

Readme & install

Copy the install command, review bundled files from the catalogue, and read any extended description pulled from the listing source.

Installation

Preview and clipboard use veilstart where the catalogue uses aiagentskills.

npx veilstart add skill danielmiessler/personal_ai_infrastructure --skill investigation

  • SKILL.md809 B

Overview

This skill performs ethical people-finding using parallel research agents to scour public records, social media, reverse lookups, and other open-source data. It magnifies human research speed by launching 15 specialized agents (45 concurrent search threads) to collect, correlate, and verify identity signals. The skill enforces strict public-data-only rules and provides confidence scoring for results.

How this skill works

On activation the skill launches 15 parallel research agents across multiple techniques: people-search aggregators, social media deep search, technical username enumeration, and public-record queries. Each agent runs multiple sub-searches, aggregates findings, cross-references identifiers (name, DOB, location, associates), and applies timeline and photo verification to produce confidence levels. The workflow always issues a pre-action notification and follows legal and ethical boundaries—no pretexting, no bypassing authentication, and no access to private databases.

When to use it

  • Find, locate, or reconnect with a person (old friends, classmates, former coworkers).
  • Reverse lookups for phone numbers, emails, usernames, or images.
  • Cross-platform social media searches and username correlation.
  • Public-record investigations: property, court, business, voter, professional licenses.
  • Verify identity when you need multi-source confirmation before contact.

Best practices

  • Provide as many unique identifiers as possible (full name variants, location, employer, school, approximate age).
  • Start with broad searches then narrow using timeline and associate matching to avoid false positives.
  • Require 3+ independent matching identifiers for HIGH confidence before acting.
  • Respect privacy: stop if the intent appears malicious or the subject has opted out.
  • Use the skill’s confidence scoring to decide whether to proceed or gather more evidence.

Example use cases

  • Locate a college roommate from 2005 using name, city, and alumni networks.
  • Reverse phone lookup to identify a caller and cross-check carrier and location.
  • Aggregate and verify a public social-media footprint for a professional contact.
  • Search property and business filings to confirm a subject’s current city and employer.
  • Perform username enumeration across platforms to find linked accounts and aliases.

FAQ

No. The skill uses only publicly available data and explicitly forbids login bypass, pretexting, or access to private databases.

How reliable are results?

Results include a confidence score: HIGH requires 3+ independent matches, MEDIUM requires 2 consistent identifiers, LOW indicates single-source or ambiguous matches.

What notification runs before searches?

Before any workflow runs the skill issues a voice/text notification indicating the workflow name and intended action, ensuring transparency before research begins.

Built by
VeilStrat
AI signals for GTM teams
© 2026 VeilStrat. All rights reserved.All systems operational