Asana patches MCP server vulnerability exposing user data
Asana discovered and patched a vulnerability in its Model Context Protocol server that could have allowed users to access data from other organizations. The company disabled the feature for nearly two weeks to apply the fix and reset all connections, though no confirmation of actual unauthorized access was provided.
What happened
Asana discovered a vulnerability in its Model Context Protocol (MCP) server implementation on June 4, 2025. The flaw could have allowed users to view data from other organizations through the AI integration feature.
Asana disabled the MCP server between June 5 and June 17 to apply the patch. The company reset all connections to the server and required users who had previously enabled the integration to reconnect.
Reported impact
The vulnerability could have potentially exposed certain information from Asana domains to other Asana MCP users. Asana did not specify how many customers were affected or whether any data was actually viewed by unauthorized parties.
Index analysis
Asana launched the MCP integration on May 1 to allow customers to query project data using natural language and third-party AI apps. The framework enables AI systems to interact with external data sources such as enterprise applications.
Governance controls
Cybersecurity firm UpGuard recommended enforcing strict tenant isolation and least-privilege access to limit data access by AI systems. It also advised organizations to log all LLM-generated queries to support future audits and investigations.
Classification
Sources and evidence
Sources not yet published
This record was researched and written by the Index from original reporting. Its sources are still being verified by an editor, and they are listed here once that check is complete. Until then the record carries no reviewer.