Block Deploys Goose Agent To Orchestrate Tools
Block says its open-source AI agent framework goose now serves as an internal orchestration layer, with roughly 60% of employees (about 6,000 people) using it weekly to connect MCP-powered extensions like Slack, Snowflake, Google Drive and Databricks. Block collaborated with Anthropic on the Model Context Protocol to standardize integrations and emphasizes guardrails, legal approvals, and curated extensions to secure sensitive data while speeding workflows.
Key Points
- 1Deploys goose across company, with 60% (~6,000) employees using it weekly for tool orchestration
- 2Establishes MCP-based integrations enabling fast cross-tool data synthesis and reduced handoffs and cycle times
- 3Requires legal approval, curated extensions, and built-in boundaries, guiding secure agent deployment practices
Scoring Rationale
Demonstrates significant enterprise adoption and open-standard collaboration, but limited novelty beyond internal implementation and incremental tooling improvements.
Sources
Public references used for this report.
Practice interview problems based on real data
1,625 SQL & Python problems across 15 industry datasets — the exact type of data you work with.
Try 250 free problems
