iDenfy launches MCP server for AI coding assistants

iDenfy has released a Model Context Protocol server that enables AI coding assistants such as ChatGPT, Claude, and Cursor to ingest the company's live API documentation in real time, eliminating the friction of hallucinated parameters and outdated endpoint references that plague automated integration workflows, according to Biometric Update. Domantas Ciulde, iDenfy's chief executive, framed the move as a response to developer fatigue, stating that "developers don't want to read fifty pages of documentation to find out the name of one field." The read-only server serves only public technical specifications without accessing customer data or executing API calls, ensuring developers retain full control over what they expose to their chosen AI interface. The launch arrives amid broader industry grappling with agentic authorization challenges. At the recent MCP Dev Summit, Gluu's Michael Schwartz advocated for context-aware policies that transcend static role-based models, while Pindrop, in commentary to NIST, warned that traditional credential checks disintegrate when AI systems act autonomously on a user's behalf. Pindrop argues that enterprises must now treat human approvals as first-class identity events, maintaining clear provenance chains and incorporating real-time liveness detection to verify that delegated actions remain authentic and attributable.