agentsclimarketplace
Sign in

Managed model endpoints

Skill xuzhougeng/wisp-science/skills/managed-model-endpoints

Explain Wisp's current managed-model endpoint boundary and plan a safe integration. Use when the user asks to register, start, stop, tunnel, authenticate, or manage a persistent inference service.From its SKILL.md

Install
npx -y skills add xuzhougeng/wisp-science --skill managed-model-endpoints

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

What its file declares

Copied from the file, not written here

The file declares its own license as Apache-2.0. That is the author’s claim about this one file, and it is not the same thing as the license GitHub reports for the repository, which is listed with the other numbers below.

SKILL.md

1.4 KB, 218 tokens by cl100k_base, as published. Nobody here has run it

Managed model endpoint boundary

Wisp does not currently expose an endpoint registry or a service-lifecycle backend. The Agent cannot allocate ports, configure tunnels, read secrets, register health checks, or start and stop a persistent inference service through Python.

Do not model service startup as a normal run_in_context command: a Run tracks one process lifecycle, while a managed endpoint also needs a durable endpoint identity, health, routing, authentication, restart policy, and ownership.

Supported path

If the user already operates an endpoint outside Wisp and the selected local, WSL, or SSH context can reach it using credentials already configured in that execution environment, load using-model-endpoint to run a bounded inference client. Never request or print secret values merely to make the call.

Otherwise explain that endpoint registration and service management are not available in this Wisp build. A future implementation should add a typed service or execution-context backend with keyring-backed secret binding, health checks, start/stop/recovery semantics, and auditable invocation Runs.

What ships with it

Read from the repository

Just SKILL.md. No reference files, no scripts.

Keep looking

Skills are one crate of 325,949. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.