GoKawiilsafe-upgrade agent combines Jev, LangGraph and Tenuo for scoped dependency updates
A team has built safe-upgrade, a dependency-upgrade agent that separates three concerns: Jev converts release notes and repository evidence into typed, probabilistic decisions; LangGraph runs those decisions through a durable workflow; and Tenuo issues each resulting action a task-scoped warrant limiting its authority. The project uses dependency upgrades as a test case because passing installs and CI checks can still hide broken APIs or behavior changes that existing tests don't catch.
GoKawiil's interpretation of the reporting above, not reported fact.
The design suggests a broader pattern for agent systems: keeping judgment (deciding what to do), control flow (sequencing the work), and authority (what actions are actually permitted) as separate layers rather than bundling them into one opaque agent. This could make it easier to audit or constrain autonomous coding agents that have access to source code, package managers, test runners, and version control, though the approach's real-world reliability outside this dependency-upgrade case study remains unproven.
- safe-upgrade is an experimental agent for automating dependency upgrades using Jev, LangGraph, and Tenuo.
- Jev handles semantic judgment about release notes and repository compatibility; LangGraph manages workflow; Tenuo scopes each action's authority.
- The project highlights that passing tests and installs doesn't guarantee an upgrade hasn't broken untested behavior.
Source: tenuo.ai — Jev In Practice, 2026-09-23
Published there as: “Jev in practice: typed decisions, scoped authority”
Read the original report → The summary and analysis above are GoKawiil's own, written from reporting by the source above. Facts and quotes belong to the original publisher.