The following adapters are currently registered in this Covia venue:
| Adapter | Class | Description |
|---|---|---|
| jvm | JVMAdapter | Provides utility functions executed on the venue's Java Virtual Machine. Supports string concatenation, URL encoding/decoding, and other common. Ideal for data processing, text manipulation, and integration with Java-based systems and libraries. |
| schema | SchemaAdapter | JSON Schema validation, inference, and coercion. Validates data against schemas, infers schemas from example data, and coerces values to match target types. Pure JVM operations. |
| agent | AgentAdapter | Manages agent lifecycle: create agents, submit requests, deliver messages, and run their transition loop. Agents are per-user, identified by human-readable names, with persistent state in the lattice. |
| test | TestAdapter | Provides various test operations for development and debugging. Supports echo operations, random data generation, error simulation, delay operations, and never-completing tasks. Perfect for testing async behavior, error handling, and orchestration workflows in the Covia platform. |
| covia | CoviaAdapter | Provides native access to internal services in this Covia venue, including lattice read/write/list/delete for user data. |
| auth | AuthAdapter | Authentication introspection — exposes the caller's authenticated identity as resolved by the venue. Use whoami to verify which DID the venue attributes to the current request. |
| goaltree | GoalTreeAdapter | Goal tree agent adapter — hierarchical goal decomposition with subgoal/complete/fail/compact harness tools. |
| secret | SecretAdapter | Manages encrypted secrets in the caller's per-user secret store. Supports storing secrets with automatic redaction in job records. |
| ucan | UCANAdapter | UCAN token operations for capability-based authorisation. |
| orchestrator | Orchestrator | Enables complex multi-step orchestration operations with dependency management and parallel execution. Supports sophisticated job orchestration with step dependencies, result aggregation, and error handling across multiple operations. Perfect for building complex AI workflows, data processing pipelines, and multi-service integrations with intelligent task coordination. |
| a2a | A2AAdapter | Outbound Agent-to-Agent client. Invoke remote A2A agents as grid operations: fetch agent cards, send messages, get/cancel remote tasks. |
| file | FileAdapter | Filesystem access for agents over a uniform tool surface. Reads, writes, lists, and manages files within operator-configured named roots. Each root can be backed by a host directory, an ephemeral temp dir (auto-cleaned on JVM exit), or a DLFS drive (lattice-backed, per-user). Agents address files by root name + relative path regardless of backend. With no roots configured the venue defaults to a single ephemeral 'tmp' root. |
| langchain | LangChainAdapter | Connects to LangChain for advanced language model interactions. Provides seamless access to local and remote AI models with configurable parameters and system prompts. Ideal for natural language processing, AI-powered conversations, and intelligent content generation workflows. |
| grid | GridAdapter | Enables distributed processing and resource sharing across the Covia network via grid operations. Provides access to remote venues, distributed job execution, and collaborative computing capabilities. Perfect for scaling computational tasks, leveraging distributed resources, and building resilient, distributed AI applications. |
| convex | ConvexAdapter | Enables interactions with the Convex network, including on-chain CVM queries and transactions |
| http | HTTPAdapter | HTTP client enables seamless web API integration and external service communication. Supports GET, POST, and other HTTP methods with custom headers, query parameters, and request bodies. Perfect for integrating with REST APIs, web services, and external data sources like Google Search and AI model APIs. |
| json | JSONAdapter | Pure data manipulation primitives over JSON / CVM data. Provides deep merge, conditional selection, path-based assoc, and key-based selection. All operations are pure functions with no IO and sub-millisecond execution. Use these to compose structured outputs and branch on prior step results in declarative orchestrations. |
| dlfs | DLFSAdapter | Decentralised Lattice File System — self-sovereign file storage with CRDT merge semantics. Manage per-user drives, read and write files, list directories. DLFS drives exist as an independent lattice region signed by the user's own key, enabling private, portable health vaults and document storage. |
| mcp | MCPAdapter | A Model Context Protocol (MCP) adapter that enables seamless integration with MCP-compatible AI models and tools. Provides standardised communication protocols for AI agents to interact with external systems and services. Essential for building sophisticated AI workflows and connecting with modern AI development ecosystems. |
| asset | AssetAdapter | Store, retrieve, and list content-addressed assets. Assets are immutable, identified by the CAD3 hash of their metadata. |
| vault | VaultAdapter | Personal health vault — read and write files in the user's private health-vault DLFS drive. Stores health documents, lab results, and records with self-sovereign encryption. No drive name needed — always operates on the user's own health vault. |
| llmagent | LLMAgentAdapter | LLM-backed transition function for agents. Maintains conversation history in agent state, processes inbox messages as user turns, and invokes a level 3 grid operation for LLM calls. Supports tool call loops: when the LLM requests tool calls, executes them as grid operations and feeds results back until a text response is produced. Built-in tools: complete_task, fail_task (added dynamically when tasks are pending). All other tools dispatch via the grid. |