Open-Source Gateway Coral Lets AI Agents Write Their Own Tools and Policies
Developers have launched Coral, an open-source, self-hosted gateway enabling AI agents to write TypeScript programs as custom tools while maintaining strict policy controls.
Photo: Hacker News Front Page
Key points
- Coral is an open-source, self-hosted gateway enabling AI agents to write TypeScript programs as custom tools.
- The system runs agent-authored code in isolated environments, validating every underlying provider call against workspace security policies.
- Successful workflows can be saved and shared across teams, reducing duplicated token usage and agent setup time.
A new open-source and self-hosted gateway called Coral has been introduced to manage how artificial intelligence agents interact with enterprise systems. Developed out of a need to streamline AI site reliability engineering, the platform sits between agents and their operational tools, allowing agents to author short TypeScript programs to accomplish multi-step workflows. Rather than starting tasks from scratch each time, agents can save successful procedures as reusable custom tools for future operations.
Coral addresses a common limitation in traditional Model Context Protocol gateways, which typically offer an all-or-nothing approach to permissions. Instead of merely controlling what an agent can reach, Coral governs specific actions through granular allow, ask, or block policies. Every code-based tool execution runs inside a secure V8 isolate without direct network access, while the gateway evaluates every underlying call against established security rules before attaching credentials and forwarding requests.
According to the creators, the software is currently utilized by thousands of developers across more than 450 companies for tasks spanning customer support, security operations, and site reliability engineering. Workspace owners can connect existing OpenAPI documents or MCP servers, approve sensitive actions like posting to external communication channels, and share verified custom tools across their entire team to reduce token consumption and repeated work.
What to watch next: Monitor adoption rates among enterprise development teams and potential integrations with additional developer environments and LLM clients.
Source Hacker News Front Page — github.com/withcoral/coral ↗
Read next
🧠AI AnalysisPedestal AI Secures Seed Funding to Automate Logistics With Intelligent Agents
Pedestal AI has launched publicly with $4.7 million in seed backing to deploy agentic artificial intelligence aimed at streamlining logistics and supply chain operations.
🏢Big TechUnited Launch Alliance Investigates Vulcan Rocket Failure
United Launch Alliance is investigating a mission failure of its Vulcan rocket, which suffered from low booster performance shortly after liftoff. The incident raises que…
🌐Work & SocietyUS Southwest Hydropower Output Hits Record Lows Amid Persistent Drought
Hydropower generation in the US Southwest dropped 28% below the 10-year average this summer as record-low water levels in the Colorado River Basin threaten regional energ…
🏢Big TechYouTube Updates View Metrics to Include TV Co-Viewing Estimates
YouTube is incorporating estimated co-viewing data into its public view counts to better reflect group consumption on television sets.
Apple Services Experience Widespread Outage
Apple users are reporting significant disruptions across iCloud, the App Store, and other core services, despite initial discrepancies in the company's status reporting.
⚖️Regulation & PolicyOpenAI Partners with Chilean Ministry to Advance Public Sector AI
OpenAI has signed a non-binding agreement with Chile's science ministry to integrate artificial intelligence into public administration and support local startups.