india company diligence

Manual due diligence on Indian companies is often slow, prone to errors, and hampered by the Ministry of Corporate Affairs (MCA) website stability. This skill provides instant access to a local mirror of 3.74 million Indian company records for immediate KYB, vendor verification, and CIN/GSTIN analysis.

What it does

• Registry lookup retrieves full registry rows by Company Identification Number (CIN) including capital and status.
• Name-based search identifies registered entities when the exact legal name is unknown or partially provided.
• CIN decoding breaks down a CIN into its industry code, state, incorporation year, and ownership type without network access.
• GSTIN validation checks GST identification number structure and state codes offline to flag basic formatting errors.
• Provenance tracking attaches source and as-of-date stamps to every data point to ensure auditability.

How it works

1. Use company_search with a business name to find the correct legal entity and its corresponding CIN.
2. Execute company_lookup using the CIN to pull the complete registry record, including ROC location and paid-up capital.
3. Run cin_decode or gstin_validate for immediate structural checks on identification numbers.
4. Synthesize the findings, explicitly stating the status (e.g., Active vs Strike Off) and the data snapshot date.

Frameworks & tools

This skill requires the india-company-diligence MCP server, the uv package manager, and a local DuckDB mirror of the MCA registry (approx. 1.4GB on disk).

Why this beats prompting it yourself

Generic LLMs lack access to the specific, massive dataset of millions of Indian companies and often hallucinate company details or status. This skill uses a verified local mirror to ensure every piece of data is grounded in official registry records with clear provenance.

Use cases

• Verify vendor legitimacy during onboarding for procurement or credit lines.
• Distinguish between active companies and similarly-named entities that are Strike Off.
• Perform high-speed batch validation of CIN and GSTIN structures in a CRM.

Known limitations

Data is a snapshot from a mirror, not a live feed, so it excludes same-day incorporations. It does not currently cover directors, charges, or financial filing history.