person-intelligence-osint（人的智力 osint）

Person Intelligence OSINT

Conduct a comprehensive, LinkedIn-friendly, open-source intelligence investigation on any individual. The process starts with professional network reconnaissance, expands to broad-spectrum public sources, and produces a structured intelligence dossier.

Inputs

Three inputs are required. If any are missing, ask the user before proceeding.

Workflow

Execute sequentially. Do not skip steps.

Step 1: Acknowledge and Scope

Confirm the target details with the user. Create a working directory:

mkdir -p /home/ubuntu/osint/{{person_name_slug}}

Create osint_notes.md inside this directory to accumulate raw findings throughout the process.

Step 2: LinkedIn Reconnaissance

This is the foundational data source. The goal is to locate and extract the subject's canonical professional profile.

1. Use search tool (type: info) with these query patterns:

- "{{person_name}}" "{{company_name}}" site:linkedin.com - "{{person_name}}" "{{job_title}}" LinkedIn - "{{person_name}}" "{{company_name}}" LinkedIn profile

1. Use browser_navigate to visit the identified LinkedIn URL. Extract:

- Full name, headline, location - Career history (all roles, companies, dates) - Education - Skills, endorsements, recommendations count - Connection count and notable mutual connections - Any posts or articles authored

1. Save all extracted data to osint_notes.md immediately.

Step 3: Corporate Intelligence

Investigate the subject's company and any other associated entities.

1. Use search tool with:

- "{{company_name}}" Companies House (UK) or "{{company_name}}" SEC filing (US) - "{{company_name}}" corporate registration - "{{person_name}}" director OR officer "{{company_name}}"

1. For UK companies, navigate to https://find-and-update.company-information.service.gov.uk/ and search for the company. Extract:

- Company number, status, incorporation date - Registered address - Directors and officers (confirm subject's role) - Filing history (recent accounts, confirmation statements)

1. Append findings to osint_notes.md.

Step 4: Media & Public Presence

Expand to news, social media, and other public mentions.

1. Use search tool (type: news) with:

- "{{person_name}}" "{{company_name}}" - "{{person_name}}" interview OR announcement OR appointed

1. Use search tool (type: info) with:

- "{{person_name}}" site:twitter.com OR site:x.com - "{{person_name}}" site:facebook.com - "{{person_name}}" site:github.com (if technical role) - "{{person_name}}" "{{company_name}}" conference OR speaker OR panel

1. For each significant result, use browser_navigate to read the full content. Save key excerpts and URLs to osint_notes.md.

Step 5: Domain & Technical Intelligence (Conditional)

Execute this step only if the subject has a technical or digital role (CTO, developer, engineer, etc.).

1. Search for personal websites, blogs, or portfolios.
2. Search for GitHub profiles, open-source contributions, or academic papers.
3. Search for domain registrations (WHOIS) if a personal domain is found.
4. Append findings to osint_notes.md.

Step 6: Connected Systems Check (Conditional)

If the user has Gmail or Slack MCP integrations enabled, search those systems for any prior correspondence or mentions of the subject.

1. Gmail: Use gmail_search_messages with queries for the person's name and company.
2. Slack: Use slack_search_public_and_private with the person's name.
3. Append any findings to osint_notes.md.

Step 7: Synthesize and Generate Dossier

1. Read the dossier template: /home/ubuntu/skills/person-intelligence-osint/templates/dossier_template.md
2. Create the final report file: Dossier_{{person_name_slug}}.md
3. Populate the template with synthesized intelligence from osint_notes.md.
4. Ensure all sections are written in complete paragraphs, not bullet lists.
5. Include a References section with numbered citations and source URLs.
6. Include a table for career history and a table for the identity profile.

Step 8: Deliver

1. Review the dossier for completeness.
2. Use message tool (type: result) to deliver the Markdown dossier as an attachment.
3. Provide a concise summary of key findings in the message text.
4. Optionally, sync to Google Drive if the user has integration enabled.

Output Format

The final dossier MUST use the template at templates/dossier_template.md. Key requirements:

• Professional Markdown formatting
• Identity table at the top
• Career history table
• Complete paragraphs (not bullet-point lists) for analysis sections
• Numbered inline citations with a References section
• All source URLs documented

Example Usage

User says: "Research John Smith, VP Engineering at TechCorp"

Skill triggers with:

• Person Name: John Smith
• Job Title: VP Engineering
• Company Name: TechCorp

Output: A structured Markdown dossier covering LinkedIn profile, corporate filings, news mentions, social media presence, and any connected system references.