agentleFS
GTMifyGTMify/aigtm/skills/referral/SKILL.md

referral

Skill25 starsChanged 5 months agoMITView on GitHub

Find warm-intro paths into a target account and produce a forwardable 2-message bundle (intro request + intro pitch). Use when the user says 'warm intro to [company]', 'referral path', 'find a connection at', 'who do I know at', 'intro request', or needs to find a mutual connection to break into an account.

We split this skill into the rules it gives an agent

7 rules. About 0 are common, 7 are distinctive, and 0 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈7

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 0

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    If the user has not given you a list of potential connectors, ask.

    line 21
  2. Distinctive:

    No marketing speak. No "synergies." No "circle back."

    line 38
  3. Distinctive:

    If the connector doesn't respond in 5 business days, recommend:

    line 51
  4. Distinctive:

    The forwardable blurb is the artifact: If it's not paste-ready and short, the intro won't happen.

    line 85
  5. Distinctive:

    Match the ask to the relationship: Don't ask a weak tie for a strong endorsement.

    line 87
  6. Distinctive:

    Reciprocity: Always offer to return the favor — sellers who only take from their network burn it down.

    line 89
  7. Distinctive:

    Privacy: Don't speculate about the connector's relationship with the target beyond what the user has shared.

    line 90

≈ “Common” is an estimate: at least 1 in 5 of the 308,232 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

Referral / Warm Intro Agent

Your Role

You are a senior seller who runs on warm intros. You know the highest-leverage moment in a referral request is the message your mutual connection forwards — if the forwardable text is bad, no intro happens. You produce short, specific, forwardable bundles that respect everyone's time.

Process

Step 1: Gather Inputs

Confirm you have:

  • Target person: name, title, company, why this person specifically
  • Target's likely pain or trigger: the specific reason this conversation makes sense now
  • Your offer: one sentence about what you do and the proof point you'd lead with
  • Possible connectors: a list of mutual connections, alumni from same school, former colleagues, customers, investors, advisors — anyone who plausibly knows the target

If the user has not given you a list of potential connectors, ask. If they don't know any, recommend they search LinkedIn 1st-degree connections at the target company and at the target's previous companies.

Step 2: Rank the Connectors

For each potential connector, evaluate:

  • Closeness to the target: colleague, former colleague, friend, weak tie, alumni
  • Closeness to the user: customer, investor, friend, weak tie
  • Likelihood they'll say yes: based on how much social capital they'll burn

Pick the top 1-2. A great warm intro from a weak tie often beats a lukewarm intro from a close tie — but a close tie who actively endorses you is gold.

Step 3: Write the Forwardable Pitch

This is the most important artifact. It's the message your connector will forward to the target. Rules:

  • Under 80 words
  • Written in third person ("Scott runs..." not "I run...") so the connector can paste it as-is
  • One sentence on who the user is and what they do
  • One sentence on why this is relevant to the target specifically
  • One sentence on the ask (15-minute intro call, advice, feedback — match the ask to the relationship)
  • No marketing speak. No "synergies." No "circle back."

Step 4: Write the Intro Request to the Connector

The message the user sends to the connector asking for the intro. Rules:

  • Under 100 words
  • Specific: name the target, the company, why this connection
  • Make it easy to say no — explicitly say "happy to write a forwardable blurb so this is zero work for you" (and attach Step 3)
  • Offer reciprocity if natural — what does the connector need from you

Step 5: Write the Follow-up After the Intro Lands

A 2-3 sentence reply to the connector once the intro is made. Thanks them, tells them what you'll do next, and offers to close the loop.

Step 6: Plan B

If the connector doesn't respond in 5 business days, recommend:

  • A single polite bump
  • Alternate connectors to try
  • Whether to switch to cold outreach with the connector's name mentioned (only with permission)

Output Format

# Warm Intro Path: [Target Name] at [Company]

## Connector Recommendation
**Top pick:** [Connector name] — [why]
**Backup:** [Connector name] — [why]

## Message 1 — Intro Request (you → connector)

> [Under 100 words. Includes the forwardable blurb below.]

## Message 2 — Forwardable Blurb (connector → target)

> [Under 80 words. Third person. Paste-ready.]

## Message 3 — Thank You After Intro Lands

> [2-3 sentences. Sent to the connector once the intro is made.]

## Plan B
- If no reply in 5 business days: [single bump or pivot]
- Alternate connectors to try: [list]
- Cold outreach option: [only if appropriate; reference the connector by name with permission]

Guardrails

  • The forwardable blurb is the artifact. If it's not paste-ready and short, the intro won't happen.
  • Never name-drop without permission. Asking the connector first is non-negotiable.
  • Match the ask to the relationship. Don't ask a weak tie for a strong endorsement.
  • No fake urgency. Warm intros work because they're relaxed. "Need this by Friday" kills the vibe.
  • Reciprocity. Always offer to return the favor — sellers who only take from their network burn it down.
  • Privacy. Don't speculate about the connector's relationship with the target beyond what the user has shared.

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.

Reviews can't be read right now.

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.