FAQ
What password does a CAS PDF use?
Section titled “What password does a CAS PDF use?”It depends on how the statement was generated:
- Default (via MyCams / MFTransact portals): PAN + Date of Birth in
DDMMMYYYYformat, e.g.ABCDE1234F01Jan1990. - Custom (generated directly from CAMS / KFintech websites): Users can set their own password during CAS generation.
The API accepts any password via the password field. If you are building a user-facing flow, let the user enter their password rather than assuming a format.
What statement types and registrars are supported?
Section titled “What statement types and registrars are supported?”CAMS and KFintech Consolidated Account Statements — both detailed and summary formats, including password-protected PDFs. These are the two registrars that handle all Indian mutual fund statements.
Should I use the REST API or MCP?
Section titled “Should I use the REST API or MCP?”Use the REST API when your product owns the workflow and just needs structured CAS output. Use MCP when your users work inside Claude, Cursor, Copilot, or another MCP-capable AI client and you want Portfolio Intel available as a tool.
How does pricing work for test and live keys?
Section titled “How does pricing work for test and live keys?”Test keys (sk_test_*) are free with rate limits — they can parse real PDFs and are safe for development. Use pdf_ref: "sample_ref" with a test key to get sample data without needing a real PDF. Live keys (sk_live_*) are metered per parse. See the Pricing page for current rates.
Can I use the docs from AI tools directly?
Section titled “Can I use the docs from AI tools directly?”Yes. A docs MCP endpoint at docs.portfoliointel.co.in/mcp serves the same documentation content to AI clients. Add it as an MCP server in any compatible client (Claude Desktop, Cursor, VS Code Copilot).
Where is data processed and stored?
Section titled “Where is data processed and stored?”All data is processed and stored in India. The architecture is SEBI-compliant with no cross-border data transfers. PDF content is never retained after parsing is complete.