| Filename | Latest commit message | Latest commit date |
|---|---|---|
b14ae9e34b
mara chose on the PR to test a reviewer-archetype agent named spawn first and leave a custom root agent (ruth) for later. So the test creates spawn with the archetype's default permissions and drops manage_root_agent, which only ruth carried. - hive-meta-flake/testdata/meta-flake-swarm.nix.in: the agent is spawn, and its web port is 8316, the port the name hashes to (ruth's was 8232). - hive-meta-flake/src/lib.rs: the unit test that compares this fixture with the rendered flake renders spawn instead of ruth. - nix/vm-tests/full-setup.nix: the agents list and step 8 use spawn. - The `swarmctl setup status` expectation keeps its "9 ruth created and up" row: that is the product's fixed step text (swarmctl/src/setup.rs), not the test's agent. The test creates no root agent, so step 9 is expected to be not done, and the subtest's comment says so. |
||
| .. | ||
| src | ||
| testdata | ||
| Cargo.toml | ||
| README.md | ||
hive-meta-flake
The meta flake's flake.nix text: one nixosConfigurations.<agent> per agent,
each built from hyperhive's agent-base plus the agent's config input or
archetype.
Pure string rendering, with no IO and no environment reads. The caller passes
the hyperhive pin, the agents with their pins, and the values every agent
container gets. swarm-controller calls render_flake and reads the pins of the
flake it renders (pinned_rev). hive-c0re uses only the agent spec and port
helpers.
testdata/meta-flake.nix.in is the render for one archetype agent with a
placeholder for the hyperhive pin. A unit test keeps it equal to render_flake's
output, and the swarm-builder-offline VM test (nix/vm-tests/) serves it.
testdata/meta-flake-swarm.nix.in is the render for two archetype agents with
the forwarded values of a single-host swarm as @<NAME>@ placeholders, kept equal
by a unit test the same way. The full-setup VM test fills it in with the
controller's own settings and builds the agents' toplevels from it.