Exa lead generation
Exa web search as standalone Agent Skills — no MCP server, just an EXA_API_KEY. Company, people, papers, code, finance, personal-site & X search + lead generation for Claude Code, Codex, Cursor, OpenCode, Gemini CLI, and any SKILL.md agent.
npx -y skills add CodeAlive-AI/exa-skills --skill exa-lead-generationAssembled from the repository path, not quoted from the project. Check it against their README if it does not work.
2 things to look at
- no licenseNo license file was found in the repository. Code published without one is not open source by default, so using it at work is a question for whoever answers licensing questions where you are.
- 2 stars2 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.
What its author says it does
Copied from the file, not written here
Build prospect lists with Exa — find target companies matching an ICP, then their decision-makers, then enrich with signals. Use for outbound sales lists, account mapping, and B2B prospecting. Multi-pass orchestration over the Exa API; no MCP server required.
The file declares its own license as MIT. That is the author’s claim about this one file, and it is not the same thing as the license GitHub reports for the repository, which is listed with the other numbers below.
SKILL.md
3.0 KB, as published. Nobody here has run it
Lead Generation (Exa)
Turn an ICP into a prospect list: companies → decision-makers → enrichment. This is a multi-pass workflow built on exa-company-research and exa-people-research. Calls the Exa REST API through local scripts; no MCP server needed, only an EXA_API_KEY.
Setup (once)
export EXA_API_KEY=your-key # or EXA_API_KEY=... in a .env at the repo root
Get a key at https://dashboard.exa.ai/api-keys. Shared details: exa-native-base.
The pipeline
Pass 1 — Source companies (this skill's script):
python scripts/lead_generation.py "Series A B2B SaaS companies in fintech" -n 20
python scripts/lead_generation.py "category:company developer tools startups hiring backend 2026" -n 20
Run 2-3 angles (category, stage, recent-launch) and dedupe to a clean company list.
Pass 2 — Find decision-makers at each company (use exa-people-research, batch 3-5 companies per subagent):
python ../exa-people-research/scripts/people_research.py "VP Engineering at <Company>" -n 5
python ../exa-people-research/scripts/people_research.py "Head of Sales at <Company>" -n 5
Pass 3 — Enrich (funding, hiring, news signals → timing):
python ../exa-company-research/scripts/company_research.py "<Company> funding round hiring news" -c news -n 5
Token isolation (essential here)
This is a fan-out task. One subagent per batch of seeds (3-5 companies each), each running Passes 2-3 and returning a compact table. Then compile, dedupe by company + person, and rank. Never run the whole enrichment loop in your main context. See exa-native-base for the orchestration pattern.
Qualifying & enriching rows
- Define what makes a valid lead before searching (stage, geo, headcount, tech stack, buying signal).
- Capture a consistent schema per row: company · website · why-it-fits · contact name · role · profile URL · signal (hiring/funding/launch) · source.
- Apply hard filters first (stage, geo), then soft filters (genuine fit). See
references/filtering.mdinexa-native-base.
Finding hidden connections
For "who are X's customers / partners", direct queries return articles, not connections. Use indirect signals:
python ../exa-company-research/scripts/company_research.py "<Company> case study customer success story" -n 5
python ../exa-native-base/scripts/exa.py contents https://company.com/customers https://company.com/case-studies --text
After you get results
Deliver a deduped table (company · fit · contact · role · profile · signal · source), lead with the highest-signal accounts, and note coverage gaps. Treat results as similarity, not validation — verify before outreach.