Key points
- The Agentic AI Foundation now maintains 10 open-source AI projects, up from three at launch.
- OpenAI donated Open Responses, a standardized API template to ease switching between model providers.
- Anthropic contributed Agent Skills, a format for packaging resources that improve AI agent output quality.
The Agentic AI Foundation has expanded its portfolio of open-source artificial intelligence tools with four new project donations. According to SiliconANGLE, the consortium, which is backed by the Linux Foundation, now oversees 10 projects in total. This increase follows its launch last December, when it managed only three initiatives. The new contributions come from major technology firms including OpenAI Group PBC, Anthropic PBC, Hugging Face Inc. and Alibaba Group Holding Ltd. These additions aim to standardize how developers interact with large language models and manage agent capabilities.
By the numbers
- 10 total projects maintained by the Agentic AI Foundation after the latest donations.
- 4 new open-source projects added to the consortium's portfolio in this announcement.
- 3 projects managed by the foundation when it was launched by the Linux Foundation last December.
What the figures mean
The growth in project count signals a push toward interoperability in the AI ecosystem. Developers currently face friction when switching between large language model providers because each uses a different application programming interface. OpenAI’s donation, Open Responses, addresses this by providing a standardized template for API implementation. This allows code to move between compatible APIs without extensive rework. Since OpenAI modeled this after its own API, existing workloads can migrate more easily, leveraging developer familiarity with that interface.
Anthropic’s contribution, Agent Skills, focuses on improving the quality of AI agent outputs. Donated in December, this project offers a standardized format for "skills," which are collections of resources like instructions, examples and automation scripts. By packaging these resources consistently, development teams can share them without restructuring files. This uniformity also helps AI agents process data more effectively, which lowers the risk of hallucinations. Anthropic engineer David Soria Parra noted the foundation provides a neutral home for such tools, including the Model Context Protocol.
What should you do?
Managed service providers should evaluate how these standardized formats affect their current AI integration strategies. Teams usually need to assess whether existing codebases can leverage Open Responses to reduce vendor lock-in. Check whether internal development workflows can adopt the Agent Skills format to streamline resource sharing. Verify that security protocols account for the increased interoperability these tools enable. Ensure that compliance teams are aware of the data handling implications of using shared skill packages.
Operational impact
Organizations should ask their engineering leads whether adopting standardized API templates reduces long-term maintenance costs. Suppliers need to confirm if their platforms support the new Agent Skills format to ensure seamless resource sharing. Internal stakeholders must understand how these changes affect vendor dependency and migration paths. Risk managers should evaluate if the reduced friction between providers introduces new exposure vectors. Communication teams should prepare updates on how these tools improve agent reliability and lower hallucination risks.
Background: Large language models
The difference between large language models and small language models lies in their parameter counts, training data volumes, and memory footprints. Large models handle complex, open-ended reasoning across many domains, while small models execute narrow tasks efficiently on local hardware with predictable latency and lower resource overhead.
What to do and how to stay safe: Agentic AI Foundation
- Review current API integrations to determine if migrating to Open Responses reduces refactoring effort.
- Assess whether the Agent Skills format aligns with existing data governance and security policies.
- Monitor the Agentic AI Foundation for updates on the Model Context Protocol and other neutral tools.
- Consult with development teams to prioritize adoption of standardized formats that improve portability.
Step-by-step guide: Shadow AI Explained: 8 Questions Answered for Secure Operations
General security guidance from the BitShield News newsroom. It is not confirmed advice from the organisations named in this story.
Frequently asked questions
Which companies donated projects to the Agentic AI Foundation?
OpenAI Group PBC, Anthropic PBC, Hugging Face Inc. and Alibaba Group Holding Ltd. donated the new projects.
What does the Open Responses project do?
It provides a standardized template for implementing AI model APIs, allowing code to move between providers with fewer changes.
How does Agent Skills help AI agents?
It offers a standardized format for packaging resources like instructions and scripts, making them easier to share and process.
