API Orchestration
Orthogonal
Trusted skills and APIs marketplace for agent discovery and execution
Agent action authorization with mandates, policy checks, human approvals, and an audit ledger
Overview
This summary uses submitted and public product material. Only source-mapped claims are documentation reviewed.
Provenant evaluates routed agent actions against explicit mandates and policies. Decisions allow an action, hold it for human approval, or deny it, with a reason and an audit record. Budgets provide an additional control for actions with a declared value.
SDK authorization calls require the integrating application to honor the decision. Gateway enforcement requires the credential and action-description boundaries documented above.
Listing decision
Agents, their mandates, and proposed actions are core product entities. The service evaluates what an agent is authorized to do and can mediate downstream execution.
Before you choose
Agent builders who need reusable authorization and human approval around actions such as sending messages, changing records, and calling business APIs.
Controls apply to integrated actions. For gateway enforcement, independent downstream credentials must be kept out of the agent runtime, and a trusted integration must map business actions to policy descriptors. The control plane is commercially licensed; Apache-2.0 licensing applies to the SDKs and examples.
Sources
First-party unless marked otherwise. Each link supports only its adjacent claim.
Documents agents, mandates, policies, approval decisions, budgets and ledger records as the governance model.
official documentation · accessed Sep 11, 2026Documents SDK and API integration, MCP tool calls, and gateway execution with downstream credentials.
official documentation · accessed Sep 11, 2026Publishes Free and paid plans, with a self-hosted option.
official pricing · accessed Sep 11, 2026Maintained integration examples and Apache-2.0 licensing for the examples.
official repository · accessed Sep 11, 2026
Comments