agent-instructions
Use when writing project instructions for a coding agent (CLAUDE.md, AGENTS.md, or equivalent). Covers what belongs in them, what does not, structure, and…
Use when researching prospects or accounts before outreach. Covers qualification, finding the trigger event, identifying the right contact, and producing a brief that makes outreach specific rather than generic.
$ npx -y skills add nimadorostkar/Claude-Skills-collection --skill lead-research --agent claude-codeHow it fires
How this skill gets triggered: by you, by Claude, or both.
/lead-researchContext preview
The summary Claude sees to decide when to auto-load this skill.
Use when researching prospects or accounts before outreach. Covers qualification, finding the trigger event, identifying the right contact, and producing a brief that makes outreach specific rather than generic.
name: lead-research description: Use when researching prospects or accounts before outreach. Covers qualification, finding the trigger event, identifying the right contact, and producing a brief that makes outreach specific rather than generic. metadata: category: business version: 1.0.0 tags: [sales, prospecting, research, qualification, outreach]
Research an account well enough that the outreach is obviously not a template. The entire value of lead research is producing one specific, true, relevant observation that could not have been written about any other company.
1. **Qualify before researching** — Does this account match the profile at all? Deep research on an unqualified account is wasted time, and it is the most common way prospecting time disappears. 2. **Find the trigger** — What changed recently? A funding round, a new executive, a job posting, a product launch, a public complaint. Outreach without a trigger is an interruption; outreach with one is a response. 3. **Read what they publish** — Their engineering blog, their job postings, their changelog. Job postings in particular are extraordinarily informative: they say exactly what a company is building and what it lacks. 4. **Find the person with the problem** — Not the most senior person. The person who owns the pain is the one who will reply. 5. **Find one specific, true observation** — Something that could not be written about any other company. This is the entire output; everything else is preparation for it. 6. **Stop when you have it** — More research does not improve outreach past this point.
**A brief that makes the outreach write itself:**
## Acme Logistics — qualification: STRONG **Profile match** - 340 employees (target: 100-1,000) - Series C, $60M, 8 months ago - Self-reported "50+ microservices" (engineering blog, March) - No dedicated platform team (inferred: see hiring, below) **Trigger — this is the reason to reach out now** Posted 4 roles in the last 3 weeks: 2x Platform Engineer, 1x SRE, 1x "Developer Experience Engineer" (a newly created role — the posting says so). The SRE posting says, verbatim: *"Our deploy pipeline takes 45 minutes and teams have started deploying less often as a result. You'll own fixing that."* They have publicly described the exact problem we solve, in a job posting, and stated that it is changing their engineers' behavior. **The person** Not the VP Engineering. The hiring manager on the DevEx posting is Priya S., Director of Engineering — she owns this problem, she is being asked to fix it, and she has just been given headcount to do it. She is the one who will reply. **The observation for the outreach** "Your SRE posting says deploys take 45 minutes and teams have started deploying less often. That second part is the expensive one — and it's usually the test suite, not the build." That sentence could not have been written about any other company. It quotes their own words back at them, and it demonstrates that we understand the problem rather than the keyword. **Disqualifiers checked** - Not already a customer - No existing vendor in this category (no relevant job-posting requirements, no mentions in their blog) - Budget: they just raised, and they have opened four requisitions.
A curated library of 137 production-grade skills for Claude and other AI coding agents. Every skill follows one structure, speaks with one voice, and earns its place by changing what the agent does.
Repo: nimadorostkar/Claude-Skills-collection
Use when writing project instructions for a coding agent (CLAUDE.md, AGENTS.md, or equivalent). Covers what belongs in them, what does not, structure, and…
Use when an agent needs state that survives a session or a context compaction. Covers what to persist, file-based memory, structuring notes for retrieval, and…
Use when automating agent behavior with lifecycle hooks. Covers hook events, deterministic enforcement of rules the model should not be trusted to remember,…
Use when packaging skills, commands, hooks, and MCP servers into a distributable plugin. Covers manifest structure, bundling, versioning, testing, and…
Use when writing a new skill for an AI agent. Covers scoping, description writing for reliable triggering, progressive disclosure, and the difference between a…
Use when reviewing or improving an existing agent skill. Covers triggering accuracy, content quality, redundancy with the base model, and measuring whether the…