agentleFS
Faaz17Faaz17/Agents-and-Skills/skills/outreach/find-contact/SKILL.md

find-contact

Skill1 starsChanged 2 months agoMITView on GitHub

Find a named human to email at a target company plus a verifiable address, and rate the confidence honestly. Use after a company dossier exists and before any draft is written. Also use when a first contact went nowhere and a different person is being considered.

The file

We split this skill into the rules it gives an agent

4 rules. About 0 are common, 3 are distinctive, and 1 is very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈3

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 1

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Never fabricate a name, a title or an address.

    line 54
  2. Distinctive:

    Never present a pattern-guess as found.

    line 55
  3. Distinctive:

    No scraping, no bulk enrichment tools, no paid address databases.

    line 57
  4. Very specific:

    If the only reachable route is careers@, say so — that is a signal to use the company's formal process and write the email differently.

    line 59

≈ “Common” is an estimate: at least 1 in 5 of the 372,429 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

/find-contact, find a real human

Usage: /find-contact Space42

The most useful output of this skill is sometimes "no verifiable address exists, here is the better route instead". A guessed address bounces, and a bounce from a personal Gmail damages every future email to that domain.

Steps

  1. Read the dossier at shared/companies/<slug>.md. It already names the team worth targeting.
  2. Delegate to the contact_finder agent. Its seniority ladder, source list and rating rules are the method.
  3. Target the right level. An engineering manager or team lead on the relevant team, first. A senior engineer on that team, second — they cannot hire but they forward, and internal forwards work. A head of engineering only at small companies. careers@ last, and expect nothing from it.
  4. Verify the address against a published source. Papers with correspondence addresses, conference speaker pages, the company's own team page, GitHub, personal sites. Record the URL.
  5. Rate it, and do not upgrade the rating hopefully. verified means seen published. pattern means guessed from a company convention. unverified means guessed. Only verified may be sent to.
  6. Check the person still works there, and record the date of the evidence.
  7. When nothing verifies, name the best alternative — LinkedIn message, contact form, or an introduction through a mutual connection — and say why the direct route failed.
  8. Append to the dossier in the format in agents/contact_finder.md, and update the person's tracker.md row.

Output format

## Contact found: <company>

**Primary**
<name> · <title> (as of YYYY-MM) · <email>
Verification: verified | pattern | unverified
Source: <URL>
Still there: <evidence, date>
Why them: <one line>

**Backup:** <name and details, or none found>

**If unverified**
Best route: <alternative>
Why the direct route failed: <one line>

**Sendable?** yes | no — <reason>
**Next:** /cold-email <person> <company>

Rules

  • Never fabricate a name, a title or an address.
  • Never present a pattern-guess as found.
  • One primary contact per company. A list of five people is a spray plan, not a contact.
  • No scraping, no bulk enrichment tools, no paid address databases. Individually found, publicly published sources only.
  • Addresses stay in this repo. Never into the Hermes vault, never into a public remote.
  • If the only reachable route is careers@, say so — that is a signal to use the company's formal process and write the email differently.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.