Cala Skill
Cala pre-processes the world's public information and serves it as structured data for AI agents. Instead of scraping web pages, your agent queries Cala and gets back typed, sourced JSON — for any real-world entity.
What Cala covers
- Organizations — companies, startups, investment firms, educational institutions, stock exchanges. Query by sector, location, funding stage, employee count, revenue, or any combination. Traverse ownership chains, subsidiaries, and parent companies.
- People — founders, executives (CEO, CTO, COO), board members, investors. Find who runs a company, who sits on a board, or which people have moved between organizations.
- Funding & investment — funding rounds by series, amount, date, and investor. Portfolio lookups for VC and PE firms. Investor co-investment patterns.
- M&A & corporate events — acquisitions, mergers, divestitures, SEC filings, and other corporate actions with sourced dates.
- Industries & markets — sector-level data, competitive landscapes, and cross-industry relationships.
- Laws & regulations — legislation and regulatory frameworks (GDPR, EU AI Act, etc.) with scope, definitions, and requirements.
- Countries & regions — geopolitical entities, economic profiles, sovereign funds, and country-level relationships.
- Products — named products, their makers, launch dates, and entity relationships.
- Events & awards — conferences, summits, industry awards, and their participants or recipients.
All responses include sources with URLs and data freshness dates.
Prerequisites
You need a Cala API key to use this skill: https://console.cala.ai/api-keys
Installing the skill
The skill file (SKILL.md) gives your agent the knowledge it needs to query Cala effectively — what tools exist, how to use them, and how to handle edge cases.
npx skills add cala-ai/cala-skill
This installs the skill for Claude Code, Cursor, Copilot, Amp, Codex, and more in one step.
Connecting to Cala
The skill alone gives your agent the instructions — pairing it with a live Cala connection lets your agent actually execute queries in real time. This combination is the recommended setup.
MCP (recommended):
For clients with native HTTP MCP support (Cursor, VS Code):
{
"mcpServers": {
"Cala": {
"url": "https://api.cala.ai/mcp/",
"headers": {
"X-API-KEY": "YOUR_CALA_API_KEY"
}
}
}
}
For Claude Desktop and other clients:
{
"mcpServers": {
"Cala": {
"command": "npx",
"args": [
"mcp-remote",
"https://api.cala.ai/mcp/",
"--header",
"X-API-KEY: YOUR_CALA_API_KEY"
]
}
}
}
Full setup guide: https://docs.cala.ai/integrations/mcp
REST API:
Base URL: https://api.cala.ai/v1
Auth header: X-API-KEY: <your-key>
Full spec: https://api.cala.ai/openapi.json
Tools
Once connected, your agent has access to five tools:
| Tool | Use |
|---|---|
knowledge_query |
Filter and list entities by criteria |
knowledge_search |
Open-ended questions with cited answers |
entity_search |
Fuzzy name → UUID lookup |
retrieve_entity |
UUID → full structured profile |
entity_introspection |
Discover available fields for an entity |
Links
- Homepage: https://cala.ai
- Docs: https://docs.cala.ai
- MCP setup: https://docs.cala.ai/integrations/mcp
- API console: https://console.cala.ai
No comments yet
Be the first to share your take.