agentleFS
Maudeunfledged834Maudeunfledged834/startup-founder-skills/skills/sourcing-outreach/SKILL.md

sourcing-outreach

Skill8 starsChanged 8 months agoMITView on GitHub

When the user needs to write recruiting outreach messages to attract passive candidates or request referrals.

1 other project uses a version of this.

Adopted by

Most recently taken up by

The file

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:

    Every outreach message should contain three elements in roughly this order:

    line 33
  2. Distinctive:

    Do not connect-and-pitch simultaneously.

    line 43
  3. Distinctive:

    Keep body under 150 words.

    line 47
  4. Distinctive:

    Always give the connector an out: "No pressure at all if this doesn't feel right."

    line 54
  5. Distinctive:

    Before writing, look for:

    line 57
  6. Distinctive:

    Follow-up 1 (Day 4-5): Add a new piece of information — a recent company milestone, a blog post from the team, or a specific project they'd …

    line 66
  7. Distinctive:

    Overselling: Don't describe the role as "once in a lifetime." Be honest and specific about what makes it compelling.

    line 75

≈ “Common” is an estimate: at least 1 in 5 of the 364,980 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

Sourcing Outreach

When to Use

Activate when the user asks to write cold outreach to potential candidates (LinkedIn InMails, cold emails), craft referral request messages, build a multi-touch follow-up sequence, or improve response rates on existing recruiting outreach. Also activate when the user is sourcing for a specific role and needs help personalizing messages at scale.

Context Required

  • From startup-context: Company name, one-line mission, stage/funding, notable investors or customers, recent milestones, and team size.
  • From user: Role being hired for, the candidate's name and background (LinkedIn profile, blog posts, talks, open-source work), what specifically drew the user to this candidate, and the communication channel (LinkedIn, email, Twitter DM).

Workflow

  1. Research the candidate — Review the candidate details provided by the user. Identify 1-2 specific, genuine connection points: a project they shipped, a talk they gave, an open-source contribution, a blog post, or a career pattern that signals fit.
  2. Choose the outreach template — Select from: cold LinkedIn InMail, cold email, warm intro request, referral ask, or follow-up. Each has different length and tone constraints.
  3. Draft the message — Write a short, personalized message using the PRC framework (see below). Keep LinkedIn InMails under 300 characters for the preview. Keep cold emails under 150 words.
  4. Add a clear, low-friction CTA — The ask should be small: a 15-minute call, a reply with interest level, or permission to send more details. Never ask for a resume or formal application in cold outreach.
  5. Build the follow-up sequence — Draft 2-3 follow-ups spaced 4-7 days apart. Each follow-up adds new information (a company milestone, a team blog post, a relevant data point) rather than just "bumping" the thread.
  6. Review for tone — Ensure the message sounds human, not templated. Check that personalization is specific enough that it could only apply to this candidate.

Output Format

  • A primary outreach message (ready to send)
  • 2-3 follow-up messages with suggested send timing
  • Notes on personalization elements used and why they were chosen

Frameworks & Best Practices

The PRC Framework

Every outreach message should contain three elements in roughly this order:

  • Personalization (P): A specific, genuine observation about the candidate's work that shows you did your homework. Not "I saw your impressive profile" — something like "Your talk on event sourcing at StrangeLoop changed how I think about our own data pipeline."
  • Relevance (R): Why this role connects to their career trajectory. Bridge from what they've done to what they'd do at your company.
  • Call-to-action (C): A single, low-commitment ask. "Would you be open to a 15-minute call this week?" is better than "Apply at our careers page."

Channel-Specific Guidelines

LinkedIn InMail:

  • Subject line matters more than body — keep it intriguing and specific (e.g., "Your Kafka work + our real-time pipeline" not "Exciting opportunity").
  • InMail preview shows ~300 characters. Front-load the personalization.
  • Do not connect-and-pitch simultaneously. Either send an InMail or send a connection request with a note — not both at once.

Cold Email:

  • Subject: Short, specific, no clickbait. "Quick question about [their project]" or "[Mutual connection] suggested I reach out."
  • Keep body under 150 words. Three short paragraphs max.
  • Plain text outperforms HTML templates. No logos, no signatures with 10 links.
  • Send from a real person's email (founder@, not recruiting@).

Warm Intro / Referral Request:

  • Make it easy for the connector: provide a forwardable blurb they can send with zero editing.
  • Include context on why you think the candidate is a fit so the connector can vouch meaningfully.
  • Always give the connector an out: "No pressure at all if this doesn't feel right."

Personalization Research Checklist

Before writing, look for:

  • Recent talks, podcast appearances, or conference presentations
  • Blog posts or technical writing
  • Open-source contributions (GitHub, GitLab)
  • Career trajectory patterns (e.g., "you've gone deep on infrastructure at two companies in a row")
  • Mutual connections or shared communities
  • Company or product they built that you genuinely admire

Follow-Up Sequence Design

  • Follow-up 1 (Day 4-5): Add a new piece of information — a recent company milestone, a blog post from the team, or a specific project they'd work on. Keep it to 2-3 sentences.
  • Follow-up 2 (Day 9-11): Try a different angle. If the first message was about the role, this one could be about the team or a technical challenge. Ask a question rather than making a pitch.
  • Follow-up 3 (Day 16-20): A graceful close. "I don't want to crowd your inbox — just wanted to leave the door open. If timing is ever better, I'd love to chat." This gets surprisingly high response rates.
  • Never send more than 3 follow-ups to a candidate who hasn't responded.

Anti-Patterns to Avoid

  • Spray and pray: Sending identical messages to 200 people. Response rates plummet below 5%. Batch personalization into tiers instead.
  • The humble brag opener: "We just raised $50M from Sequoia" as your first sentence. Lead with the candidate, not yourself.
  • Fake personalization: "I was really impressed by your background" is worse than no personalization because it signals a template.
  • Overselling: Don't describe the role as "once in a lifetime." Be honest and specific about what makes it compelling.
  • Asking for too much: "Please send your resume and three references" in a cold message guarantees no reply.

Related Skills

  • job-description — The JD provides the source material for what makes the role compelling in outreach.
  • interview-kit — Mention the candidate-friendly interview process in outreach to reduce friction.

Examples

Prompt: "Write a cold LinkedIn InMail for a senior ML engineer. She gave a talk at NeurIPS on efficient fine-tuning and currently works at Stripe."

Good output snippet:

Subject: Your NeurIPS talk on efficient fine-tuning

Hi [Name],

Your NeurIPS presentation on LoRA variants for production models stuck
with me — especially the bit about keeping inference costs flat while
scaling model complexity. We're solving a similar problem at [Company]:
building real-time ML models for [use case] and need someone who thinks
about efficiency as a first-class constraint.

We're 18 people, Series A, and this would be our second ML hire. You'd
own the model architecture end to end.

Would you be open to a 15-minute call to see if there's a fit?

— [Your name]

Prompt: "Write a referral request email I can send to a mutual connection."

Good output snippet:

Subject: Quick ask — know anyone strong in product design?

Hi [Connector],

We're hiring our first product designer at [Company] and I thought of
you since you work with strong designers regularly. Here's what we're
looking for: [1-2 sentences on the role].

If anyone comes to mind, I'd be grateful for an intro. Here's a
forwardable blurb:

---
"[Your name] is the CEO of [Company] ([one-liner]). They're hiring a
founding product designer to own the end-to-end user experience. The
team is 14 people, Series A, remote-first. Here's the JD: [link]"
---

No pressure at all — and thanks either way.

[Your name]

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.