mirror of
https://github.com/docker/docker-agent.git
synced 2026-11-04 10:02:15 +00:00
David Gageot
20b25b3da9
Remove the leading '#!/usr/bin/env docker agent run' shebang and following blank line from all example YAML and HCL files, as the config format is now self-documenting.
51 lines
1.6 KiB
YAML
51 lines
1.6 KiB
YAML
# Demonstrates the `unload` on_agent_switch builtin hook.
|
|
#
|
|
# Two agents share Docker Model Runner but use different models that don't
|
|
# fit in GPU memory at the same time. Wiring the `unload` builtin into
|
|
# each agent's `on_agent_switch` hook chain asks the previous agent's
|
|
# DMR endpoint(s) to release GPU memory every time the active agent
|
|
# transfers control. The hook is pure: it reads the model snapshot the
|
|
# runtime ships on every on_agent_switch dispatch and POSTs to DMR's
|
|
# `_unload` endpoint over plain HTTP — no provider-specific runtime
|
|
# coupling. For cloud-only providers (OpenAI, Anthropic, ...) the hook
|
|
# is a silent no-op since they don't expose an HTTP unload endpoint.
|
|
#
|
|
# Switching back and forth between `coder` and `reviewer` therefore costs
|
|
# one model load per switch instead of failing on out-of-memory.
|
|
|
|
agents:
|
|
coder:
|
|
model: qwen3-large
|
|
description: Writes Go code on demand.
|
|
instruction: |
|
|
You write idiomatic, well-tested Go.
|
|
When you finish a change, hand off to `reviewer`.
|
|
handoffs:
|
|
- reviewer
|
|
hooks:
|
|
on_agent_switch:
|
|
- type: builtin
|
|
command: unload
|
|
|
|
reviewer:
|
|
model: qwen3-coder
|
|
description: Reviews code for clarity and correctness.
|
|
instruction: |
|
|
You critique Go code written by `coder`. Be concise.
|
|
Hand back to `coder` with concrete change requests.
|
|
handoffs:
|
|
- coder
|
|
hooks:
|
|
on_agent_switch:
|
|
- type: builtin
|
|
command: unload
|
|
|
|
models:
|
|
qwen3-large:
|
|
provider: dmr
|
|
model: ai/qwen3
|
|
|
|
qwen3-coder:
|
|
provider: dmr
|
|
model: ai/smollm2
|