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

keyword-search

Skill8 starsChanged 3 months agoMITView on GitHub

Find people actively looking for your service right now using platform search. Use when the user wants to find clients on Twitter/X, LinkedIn, or Telegram by searching for hiring posts, help requests, or project announcements. Also use when the user mentions "find clients," "keyword search," "who's hiring," "looking for leads," "find prospects on Twitter," or "social selling." This is the fastest path to first clients — you're finding people who already need help today. For writing the outreach message after finding them, see cold-dm. For building prospect lists from directories and databases, see prospecting-lists.

The file

We split this skill into the rules it gives an agent

2 rules. About 0 are common, 2 are distinctive, and 0 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈2

    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:

    Check recency (3+ months old = position probably filled) and legitimacy (followers, posting history, bio — especially in crypto/Web3, …

    line 43
  2. Distinctive:

    Don't jump straight into DMs with a sales pitch.

    line 51

≈ “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

Keyword Search for Client Acquisition

Your goal is to help the user find people who are actively looking for their service right now — not in theory, not someday, but today.

Before Starting

Understand the situation (ask if not provided): what service they sell, which platforms they're active on (Twitter/X, LinkedIn, Telegram), their niche (industry, stage, geography, budget), and whether they have a portfolio link or case study ready for DMs.

The Process

Step 1: Build Your Keyword List

Think about how a potential client would describe their need in a post. Start specific, then broaden.

By service type:

  • Development: "looking for a developer," "hiring a developer," "need a dev team," "looking for a technical co-founder"
  • Consulting/fractional: "looking for a CTO," "fractional CTO," "need help with architecture"
  • Design: "hiring a designer," "need UI/UX help," "looking for a design agency"

Creative patterns that find hidden opportunities:

  • "can anyone recommend" / "does anyone know a good" + your role
  • "we are looking for" + your service
  • "struggling with" + problem you solve
  • "just raised" / "just got funded" (companies that just raised need to build)

Step 2: Search and Filter

PlatformHow
Twitter/XSearch your keyword, filter by "Latest" (not "Top") for fresh posts, check 10-30 results daily
LinkedInSearch bar with service-related keywords, filter by "Posts" + "Past week," check company pages of recently funded startups
TelegramJoin industry-relevant groups, search within groups for keywords, watch for people asking for help or announcing projects

Step 3: Qualify Before You Engage

Check recency (3+ months old = position probably filled) and legitimacy (followers, posting history, bio — especially in crypto/Web3, filter out scam accounts).

Good signals: real/funded company, clear budget signals (timeline, scope, budget mentioned), decision-maker posting (founder, CTO, VP Eng), active engagement.

Skip: vague "just an idea" posts with no budget, no-history/fake-follower accounts, posts older than 2-3 weeks, co-founder asks offering only equity.

Step 4: Warm Before You Pitch

Don't jump straight into DMs with a sales pitch. Follow them first, then leave a short, helpful comment on their post ("Hey, I might be able to help with this — sending you a DM") before moving to cold-dm for the actual message.

Daily Routine

ActivityTimeVolume
Keyword search across platforms15-20 minCheck 3-5 keyword variations
Qualify results10-15 minReview 10-30 posts
Leave comments on qualified posts10-15 min5-10 comments
Send DMs (see cold-dm skill)15-20 min5-15 personalised DMs

Total: ~60 minutes/day. Do this consistently for 30 days before judging results.

Automation (Only After Manual Mastery)

Once you've done this manually for 2-4 weeks and understand what works, automation amplifies it — don't automate before then, or you'll amplify garbage instead:

  • Drippi — automates X DM outreach at scale
  • Inbox App — multi-platform outreach automation
  • TG Finder (tgfinder.xyz) — finds Telegram handles of founders from domains
  • Custom alerts — set up Google Alerts or Twitter notifications for your keywords

Related Skills

  • cold-dm — write the actual outreach message after finding prospects
  • follow-up-sequences — what to do when they don't reply (most won't)
  • profile-optimization — make sure your profile converts when they click through
  • prospecting-lists — build prospect lists from databases, not just search

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.