Anaconda, the trusted foundation for AI-native development, today announced new capabilities across the Anaconda Platform that pair agentic development with autonomous security testing. The expansion brings agent swarms and autonomous red-team agents together with trusted packages, models, and environments to help builders ship faster and address security weaknesses before production. Builders gain greater choice in the tools and models they use, with governance built in from the start.

Anaconda’s recent survey of AI-native builders revealed strong momentum behind agent swarms with 63% of respondents moving toward the technology in some form – a clear signal that swarms are quickly becoming standard practice. As AI agents take on more builder tasks and gain access to enterprise tools and data, organizations need security testing and runtime controls that keep pace. That gap shows up in Enkrypt AI’s research, which scanned more than 268,210 agent tools across 25,264 MCP servers over four months and found vulnerabilities in 73% of them. Builders need visibility into which agent acted, what data it accessed, which tool it called, and whether it stayed within its assigned authority.

“Anyone building AI applications right now is in a tricky spot trying to balance the pressure to move fast with the growing public concern over security and safety risks. Enterprises need a way to identify and minimize risk in their AI systems before they reach production,” said David DeSanto, Anaconda CEO. “Introducing agent swarms and autonomous red-teaming agents will give our customers the ability to secure as fast as they build.”

Powered by its recent acquisitions of Kilo Code, Enkrypt AI, and Outerbounds, Anaconda helps enterprises build and strengthen their own AI Dev Factory through three new capabilities:

  • AI Workspaces brings together agentic tools and trusted components so builders and data scientists can accelerate their work, with model freedom and cost optimization.

    • Agent Swarms reach VS Code for the first time through Kilo, the primary AI workspace, dynamically coordinating multiple agents to build in parallel, share context, and optimize token costs.

    • Local AI development with Kilo Desktop brings together software engineering, data science, and secure Python environment management, with access to 500+ AI models, 19,000+ vetted packages, local model execution, and live notebook sessions where users and agents edit and run cells together.

    • Sign in with ChatGPT gives users access to every Kilo surface without extra logins or added token costs.

    • Source-built packages adds 13,000+ newly vetted AI, ML, and Python packages, extending the world’s largest source-built library.

    • Model Catalog provides access to 77 vetted open source models in a curated library, expanding builder choice without sacrificing governance.

    • MCP extends the same trust and governance to agent tool calls through the new Anaconda MCP.

  • AI Security and Guardrails deliver continuous security testing and runtime protection, helping teams catch weaknesses before release and manage risk as AI systems evolve.

    • Red-teaming challenges models, agents, and MCPs across 300+ attack categories, adapting in real time to surface weaknesses in deployment and production.

    • Guardrails expand runtime protection, approving, modifying, or blocking risky behavior across agents, tools, RAG, and MCP.

    • Agent Incident Registry gives enterprises a verified, source-backed record of publicly reported agent incidents, an industry first offering an independent way to confirm security claims beyond vendor statements.

  • AI Orchestration delivers repeatable workflows and reproducible environments, so teams can move AI reliably from development through testing into production at scale.

    • Orchestration workflows bring native access to AI Artifacts, providing governed packages, models, and security and license policies that move with each workflow from development through production.

    • Interactive inference lets teams deploy models straight from the Anaconda model catalog and query them for results, no separate deployment step required.

    • Reproducible container images are built automatically by FastBakery, which compiles conda and PyPI dependencies, including native libraries, into images pip-only tools can’t produce.

Industry Support

“We’re consistently seeing security and visibility separate enterprises that scale AI successfully from those that stall,” said Mark Beccue, Omdia Chief Analyst, AI. “Our research found that 72% of organizations rank managing this growing autonomy as critical or very important to their AI strategy, underscoring the need for security and visibility to move at the same pace as the agents themselves.”

“Kilo removed our biggest constraint. We went from being blocked by two decades of backlog to running out of tickets, and now our limit is how much one person handles at once,” said Robert Björne, Product Owner, IntraPhone. “I already run prototypes, write specs, and build things solo because Kilo lets me move that fast alone. Agent swarms are the next step, coordinating several of those workstreams at once instead of one after another, turning what one developer does in a day into what a small team does.”

To learn more about how Anaconda enables customers to unlock the power of agentic AI development safely, visit our website and register for our virtual customer conference, Anaconda Scale, live on October 15.

About Anaconda

Anaconda is the trusted foundation for enterprise AI. More than 1.2 million organizations, including 95% of the Fortune 500 and companies like Visa, American Express, SAP, UBS, and Boeing, rely on Anaconda to build and scale AI-native applications. Anaconda brings more than a decade of open source leadership to a single platform that is available across hybrid AI environments and cloud platforms such as AWS, Databricks, Snowflake and more. Visit anaconda.com or connect with us on LinkedIn.

Media gallery

About The Author
Business Wire