Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →codealive-ai/exa-company-research
Research companies with Exa semantic search — find competitors, funding, headcount, news, and build company lists.
Passed 21 of 22 test cases with the skill, and 7 of 22 without it.
and 10 more in the full report
14 fixed7 passed either way1 still failing0 broke
Tokens, turns and time describe resource use, not a dollar price.
One of 14 test cases the skill fixed. Same model, same prompt, with and without the skill.
When performing a broad web search for 'latest advancements in quantum computing hardware' using the shared client's search command, what flag should be included to ensure the full page contents are returned alongside the search results?
| Checked | Without skill | With skill |
|---|---|---|
| specifies the --text flag | Missed | Passed |
The instructions your model reads when the skill loads.
Find and profile companies using Exa's semantic search — discovery by category/stage/geography, competitor sets ("companies like X"), funding and news. Calls the Exa REST API through a local script; no MCP server needed, only an EXA_API_KEY.
export EXA_API_KEY=your-key # or put EXA_API_KEY=... in a .env at the repo root
Get a key at https://dashboard.exa.ai/api-keys. The script auto-loads a .env from the repo root, so setup is one-time. Shared details: see exa-native-base.
python scripts/company_research.py "<describe the companies you want>" [-n N] [-c CATEGORY] [--json]
Examples:
python scripts/company_research.py "AI infrastructure startups in San Francisco" -n 15
python scripts/company_research.py "companies like Stripe" -n 10
python scripts/company_research.py "Anthropic funding rounds investors" -c news --start-published 2025-01-01
Output is a Markdown list (title · date · snippet) ending in sources_reviewed: N. Add --json for raw results, --text to include page text.
Don't run bulk searches in your main context. For multi-angle research, dispatch a subagent per workstream (e.g. one for competitors, one for funding, one for news), tell each to run the script and return a compact table, then merge + deduplicate. This keeps raw search output out of your context. See exa-native-base for the orchestration pattern.
category=company returns structured company data (description, funding, headcount). Write queries that describe the company, not just a name.
# By category / geo / stage
python scripts/company_research.py "category:company developer tools for API testing" -n 10
python scripts/company_research.py "Series B fintech payments companies" -n 10
# Competitive set (layer multiple angles, then dedupe)
test cases passed, vs 7 / 22 without
Published skills whose descriptions closely match this one, with their headline benchmark lift.
More skills from the same source and nearby in the registry, with their headline benchmark lift.