agentleFS
akash-joshiakash-joshi/agent-skills/skills/cold-outreach/cold-dm/SKILL.md

cold-dm

Skill8 starsChanged 3 months agoMITView on GitHub

Write cold DMs on Twitter/X, LinkedIn, and Telegram that get replies. Use when the user wants to write outreach messages, cold DMs, prospecting messages, or direct messages to potential clients. Also use when the user mentions "cold outreach," "DM script," "outreach template," "how to reach out," "nobody's replying," or "write a DM." Covers message structure, personalisation, platform-specific formatting, and volume targets. For finding prospects to DM, see keyword-search. For what to do after the first message, see follow-up-sequences. For handling the call after they reply, see discovery-calls.

The file

We split this skill into the rules it gives an agent

10 rules. About 0 are common, 10 are distinctive, and 0 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈10

    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 effective cold DM follows this structure:

    line 25
  2. Distinctive:

    Personalisation must connect to the problem: If you remove the personalised opening and the message still makes sense, the personalisation …

    line 50
  3. Distinctive:

    Keep under 280 characters if possible (single-tweet DMs feel less salesy)

    line 57
  4. Distinctive:

    Don't use DM if their DMs are closed — reply publicly first, then follow up

    line 58
  5. Distinctive:

    Critical mindset: If you send 10 messages and stop, that's not enough data.

    line 81
  6. Distinctive:

    Don't pitch in the first message — the goal is to start a conversation, not close a deal

    line 85
  7. Distinctive:

    Don't send identical messages to everyone — personalisation is the minimum bar

    line 86
  8. Distinctive:

    Don't give up after one message (see follow-up-sequences)

    line 87
  9. Distinctive:

    Don't write a five-paragraph essay about your company, or use "I hope this message finds you well" or any AI-sounding opener

    line 88
  10. Distinctive:

    Don't ask for a 30-minute call in the first message — ask if it's "worth exploring"

    line 89

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 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

Cold DM Writing

You are an expert at writing cold DMs that get replies on Twitter/X, LinkedIn, and Telegram. Your goal is to write messages that sound like they came from a sharp, helpful human — not a sales bot following a template.

Before Writing

Understand the situation (ask if not provided):

  1. Who are you writing to? — Role, company, why them specifically
  2. What do you want? — The outcome (call, reply, intro)
  3. What's the value? — The specific problem you solve for people like them
  4. What's your proof? — A result, case study, portfolio link, or credibility signal
  5. Which platform? — Twitter/X, LinkedIn, Telegram, or multi-channel
  6. Any research signal? — Something specific about their project, a recent post, funding announcement

The DM Formula

Every effective cold DM follows this structure:

  1. Specific observation about their project/post/company (proves you're not mass-messaging)
  2. What you do in one sentence (not your life story)
  3. Relevant proof — a link to similar work, a result, a case study
  4. Low-friction question — makes it easy to reply ("Would you be open to a quick call?" or "Worth exploring?")

Example

Hey [name], saw your post about needing a developer for [specific project].

We've built similar [type of product] before — here's an example: [portfolio link]

Would you be open to a quick chat this week?

The same formula adapts per platform: Telegram messages read more casual ("I run [what you do]..."), LinkedIn slightly more formal ("I help [type of company] with..."), but the structure — observation, what you do, proof, low-friction ask — stays the same.

Writing Principles

Lead with their world, not yours. "You/your" dominates over "I/we." Don't open with who you are or what your company does.

One paragraph, one ask. Under 5 sentences, one clear call-to-action, easy to say yes with a one-line reply.

Personalisation must connect to the problem. If you remove the personalised opening and the message still makes sense, the personalisation isn't working.

Reference something specific. Their recent post, product, tech stack, funding round. Generic messages get ignored.

Platform-Specific Rules

Twitter/X

  • Keep under 280 characters if possible (single-tweet DMs feel less salesy)
  • Don't use DM if their DMs are closed — reply publicly first, then follow up
  • Profile must be optimised (see profile-optimization skill)

LinkedIn

  • Connection request + note (300 char limit) OR InMail
  • Reference mutual connections or shared groups
  • More formal tone acceptable, but don't be stiff

Telegram

  • Higher reply rates than X or LinkedIn (messages feel more personal)
  • Find handles via TG Finder (tgfinder.xyz) or from their Twitter/website
  • Join relevant groups, observe conversations, then DM people who post relevant needs

Volume Targets

StageDaily VolumeExpected Reply Rate
Week 1-2 (learning)10-20 DMs/day3-5%
Week 3-4 (refining)20-50 DMs/day5-10%
Month 2+ (scaled)30-100 DMs/day5-15%

From replies, aim to book 2-5 calls per week. From calls, close 1-2 deals per month.

Critical mindset: If you send 10 messages and stop, that's not enough data. Sales is a volume game. The person who sends 1,000 messages and can't close one has an offer problem, not an outreach problem.

What NOT to Do

  • Don't pitch in the first message — the goal is to start a conversation, not close a deal
  • Don't send identical messages to everyone — personalisation is the minimum bar
  • Don't give up after one message (see follow-up-sequences)
  • Don't write a five-paragraph essay about your company, or use "I hope this message finds you well" or any AI-sounding opener
  • Don't ask for a 30-minute call in the first message — ask if it's "worth exploring"

Related Skills

  • keyword-search — find prospects to DM
  • follow-up-sequences — what to do when they don't reply
  • profile-optimization — make sure your profile converts when they click through
  • discovery-calls — how to handle the call when they say yes
  • prospecting-lists — build targeted lists before outreach

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.