Arcade.dev secured a $60 million Series A round, led by SYN Ventures. This funding, bringing total capital to $72 million, accelerates the company’s MCP-based runtime platform for secure, governed AI agent actions in enterprise environments.
Arcade announced a $60 million Series A funding round, led by SYN Ventures, with strategic participation from Morgan Stanley and Wipro. This brings the company’s total funding to $72 million, following a $12 million seed round in 2025 led by Laude Ventures.
Arcade positions itself as the MCP (Model Context Protocol) runtime, a secure action layer for production AI agents. It addresses core barriers preventing agents from moving beyond demos: authorization (agents acting with user specific permissions without overprivileged service accounts), reliability (purpose built tools that reduce hallucinations and retries), and governance (full audit trails for every action, including which agent acted on behalf of which user in which system).
What is Arcade.dev?
Founded in 2024 in San Francisco by Alex Salazar (CEO, former Okta product leader with experience in identity and access management) and Sam Partee (CTO, former principal applied engineer at Redis with deep AI and infrastructure expertise), Arcade emerged from the need to make AI agents enterprise ready. The team includes veterans from Okta, Snowflake, Redis, Airbyte, and MongoDB. They authored the MCP tool authorization specification (adopted by Anthropic) and contribute to MCP security and governance steering committees.
MCP, an open standard pioneered by Anthropic, standardizes how AI applications connect to external systems, data sources, and tools, analogous to a “USB-C for AI.” Arcade provides the production runtime layer on top: handling dynamic user authorized actions, a catalog of 8,000+ pre built MCP tools (far more than the broader ecosystem), policy enforcement (pre and post tool call hooks), compliance features (SOC 2, SSO, RBAC, audit logs), and flexible deployment (cloud, on-prem, air gapped, hybrid).
Key differentiators include:
- User delegated authorization via existing IDPs/OAuth, avoiding shared tokens or service accounts.
- Reliable tools optimized for agent use cases rather than simple API wrappers.
- Governance and observability for security teams.
- Framework agnostic integration with LangChain, LlamaIndex, CrewAI, OpenAI Agents, Claude, etc.
Customers and testimonials highlight enterprise traction with LangChain, a top US bank, Prosus, and others. Tool call volume grew 25x in six months. Testimonials come from Harrison Chase (LangChain), and leaders at coaching, sales, and education AI companies.
The round reflects surging enterprise interest in agentic AI, moving agents from chat interfaces to autonomous actions in core systems (Google Workspace, Slack, Salesforce, etc.). Security and compliance are major blockers; Arcade solves the “who did what on behalf of whom” question that stalls deployments.
- SYN Ventures (lead): Jay Leek (Managing Partner) joins the board. Emphasis on production reality and infrastructure for safe scaling.
- Strategic investors (Morgan Stanley, Wipro): Signal enterprise validation, potential go to market synergies, and focus on regulated industries. Morgan Stanley highlighted Arcade’s authorization and governance for operational AI use.
No public valuation details were disclosed, but the jump from a modest seed to a large Series A in roughly a year indicates strong momentum and perceived market leadership in the emerging “action layer” for agents. Pricing is usage based (free tier to enterprise), scaling with tool executions and user challenges.
The AI agent market is exploding as LLMs advance, but production deployment lags due to integration, security, and reliability gaps. Arcade sits at the intersection of identity/auth infrastructure (Okta-like) and AI tool-calling platforms. By authoring key MCP specs and shipping the most production tools, it has established defensible moats in standardization and ecosystem breadth.
Competitors exist in broader AI infrastructure, gateways, or specific integrations, but Arcade claims uniqueness as the dedicated secure runtime with governance depth. Its open source contributions (e.g., arcade-mcp framework) and partnerships (Anthropic on URL Elicitation for secure flows) strengthen its position.
Use of funds: Accelerate product development (more tools, deeper governance), ecosystem growth, and hiring to support scaling from pilots to thousands of production workflows across Fortune 500 companies.
Recommended: Dapple Raises $30 Million In Seed Funding Round
Strengths:
- Timely and focused: Directly tackles the #1-3 blockers for agent adoption in enterprises.
- Strong team and pedigree: Proven infrastructure builders with relevant domain expertise.
- Traction and standards leadership: Rapid usage growth, key MCP contributions, enterprise customers.
- Strategic backing: Investors with deep enterprise networks.
- Deployment flexibility and compliance: Appeals to regulated sectors.
Potential challenges:
- Rapidly evolving MCP ecosystem and AI standards could require ongoing adaptation.
- Competition from hyperscalers, framework providers, or new entrants building similar layers.
- Execution risk in scaling sales, support, and tool catalog for diverse enterprise environments.
- Dependency on broader agent adoption; if enterprises move slower than expected, growth could moderate.
This funding validates Arcade’s thesis that the “secure action layer” will be foundational infrastructure for the agentic era, much like identity and data layers were for prior enterprise waves. With significant capital, standards influence, and early production wins, Arcade is well positioned to capture a central role as organizations deploy more autonomous AI workflows. The next 12-18 months will likely focus on deepening integrations, expanding the tool ecosystem, and proving ROI at scale.
Please email us your feedback and news tips at hello(at)dailycompanynews.com
