agentsclimarketplace

Personalized cold outreach

Skill Celina-create/X-Studio/skills/personalized-cold-outreach

A solo studio for growth operators — skills, playbooks, templates.From the repository description

Install
npx -y skills add Celina-create/X-Studio --skill personalized-cold-outreach

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

One thing to look at

  • 1 stars1 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

SKILL.md

4.0 KB, ~1.0k tokens by cl100k_base, as published. Nobody here has run it

personalized-cold-outreach

Purpose — Generate 1-to-1 personalized outreach messages at batch scale. Each message quotes the recipient's own words back to them, references a specific pain they've expressed, and offers a concrete next step.

Triggers — "personalize cold emails for <pool>", "write outreach for this CSV", "1-to-1 message for each row in <file>"

Why this beats a mailmerge — Standard mailmerge replaces {{name}}. This skill generates 3–5 personalized sentences per recipient referencing their actual public pain statements. Reply rates are typically 3–10× standard cold email.


Inputs

  • Seed CSV from competitive-intel-mining — must include handle, source_url, exact_quote, pain_category
  • Your pitch — 2–3 sentences describing what you'd show the recipient if they replied (not your full product tour)
  • Message channel — email / X DM / GitHub issue / LinkedIn
  • CTA — what action you want them to take (reply / book call / click pre-filled issue / watch 3-min Loom)

Outputs

  • A CSV enriched with columns: subject, opening, pain_mirror, pitch_line, cta, pre_filled_url (if applicable), send_batch
  • A review list of the 5 best and 5 worst generated messages for manual tuning before batch send

Steps

  1. Read row + source URL — For each recipient, pull the exact_quote and (optionally) re-open the source URL for 1–2 adjacent sentences of context.
  2. Draft opening — 1 sentence referring to the specific place you found them ("saw your Paperclip #447 thread"). Never generic ("I noticed you").
  3. Draft pain mirror — 1–2 sentences quoting their own words and naming the underlying pain ("you said 'agent panics when sandbox rejects bash' — that's the class of bug that comes from X").
  4. Draft pitch line — 1 sentence only. What you'd show them if they replied. Concrete, not abstract.
  5. Draft CTA — 1 specific next step with near-zero friction. Examples:
    • "Want a 3-min Loom tailored to your case? Reply y/n."
    • "Open a pre-filled issue with what breaks today — <URL>"
    • "Reply with a screenshot of your current agent config"
  6. Sign-off — First name only. No title, no links in signature.
  7. Quality check each message against the rules below. Regenerate if any fail.
  8. Batch output — Group into send batches of ≤10 per day per channel to avoid spam filters / rate limits.

Quality bar (per message)

  • ≤120 words total
  • Zero generic phrases ("I hope this finds you well", "I came across your profile")
  • Exact quote from their words appears verbatim once
  • Subject line is <50 chars and references something specific to them
  • CTA is a single verb + ≤15 words
  • If possible, includes a pre-filled URL so they can reply with 1 click

Message template (default)

Subject: <specific thing from their post, e.g., "The Paperclip #447 panic">

Hey {{name}},

Saw your {{source}} where you wrote: "{{exact_quote}}"

That's the class of pain that happens because {{one-line technical/strategic take}}.

I've been working on {{one-line pitch}} — here's what I'd show you if it helps:
{{pre_filled_url}}  (30 seconds, no signup)

Worth a look?

—{{your_first_name}}

Anti-patterns

  • Don't use "I think" or "I believe" — they read as low-confidence. State the observation.
  • Don't close with "let me know if you're interested" — it's a weak CTA. Use a specific verb.
  • Don't send on Monday morning — inbox is overloaded. Tuesday–Thursday 10am recipient local time performs best.
  • Don't A/B 10 variants — pick one good message, iterate after 20 replies.

Related

What ships with it

Read from the repository

Just SKILL.md. No reference files, no scripts.

Gives 0 of the 12 instructions most sales audience skills give in ~1.0k tokens

Counted across 401 of the 401 authors here whose files we hold, read 2026-08-07

  • Read product marketing context before asking questionsin 21 of 401, across 11 files
  • Acknowledge competitor strengths honestlyin 18 of 401, across 7 files
  • Start every page with a summaryin 15 of 401, across 4 files
  • Use a single, low-friction call to actionin 15 of 401, across 7 files
  • Create a single source of truth for each competitorin 14 of 401, across 3 files
  • Make each follow-up email add new valuein 11 of 401, across 5 files
  • Cut any sentence that does not drive a replyin 10 of 401, across 4 files
  • Tie personalization directly to the problemin 10 of 401, across 4 files
  • Write paragraph comparisons for each dimensionin 9 of 401, across 3 files
  • Link between related competitor pagesin 9 of 401, across 3 files
  • Keep subject lines short and lowercasein 9 of 401, across 3 files
  • Define ideal customer profile from top customersin 9 of 401, across 3 files

Said here and by no other author read

  • quote the recipient's exact words verbatim
  • write subject lines under 50 characters
  • use a single verb call to action under 15 words
  • group messages into daily send batches of 10
  • include a pre-filled URL when possible
  • sign off with first name only

Grouped from the skills themselves: near-identical wordings counted once, and counted by distinct author, so one author publishing three of these counts once. Length counted with cl100k_base; the agent that loads this file may tokenize it differently.

Keep looking

Skills are one crate of 325,949. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.