agentleFS
Sign inStart free
NVIDIAKnown publisherNVIDIA/nvcf/src/invocation-plane-services/llm-api-gateway/AGENTS.md

nvcf / llm-api-gateway

AGENTS.md261 starsChanged 9 days agoApache-2.0View on GitHub

Native Go image-source subtree for the LLM API gateway and rate-limit sync worker. Bazel is the CI build path: Follow the release workflow for source versions and producer handoffs. Account for both the main gateway and rate-limit sync worker images. The @llm-function-type scenarios in the single- and multi-cluster Helmfile features cover LLM routing and request handling. See tests/bdd/README.md to run them. - github.com/olric-data/olric is replaced by github.com/max007-008/olric on branch cas. The fork keeps the upstream module path and adds the…

The file

We split this AGENTS.md into the rules it gives an agent

3 rules. About 0 are common, 2 are distinctive, and 1 is very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈2

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 1

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  • could be automatic: a setting could enforce it
  1. Distinctive:

    Follow the release workflow for source versions and producer handoffs.

    line 28
  2. Very specific:

    If the CAS work is upstreamed, remove the replace directive and bump the upstream requirement.

    line 49
  3. Distinctive:

    Do not hand-edit generated protobufs or mocks.

    line 51

≈ “Common” is an estimate: at least 1 in 5 of the 40,246 AGENTS.md files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
AGENTS.md

AGENTS.md - llm-api-gateway

Native Go image-source subtree for the LLM API gateway and rate-limit sync worker.

Layout

  • Go module: github.com/dsx-ai-factory/nvcf/src/invocation-plane-services/llm-gateway
  • Main gateway: cmd/llm-api-gateway/main.go
  • Rate-limit sync worker: cmd/llm-api-gateway-rate-limit-sync-worker/main.go
  • Runtime config and handlers live under internal/

Build and Test

go build ./...
go test ./...
gofmt -w <changed-go-files>
kustomize build kustomize/overlays/local

Bazel is the CI build path:

bazel test //... --flaky_test_attempts=3

Follow the release workflow for source versions and producer handoffs. Account for both the main gateway and rate-limit sync worker images.

LLM routing BDD

The @llm-function-type scenarios in the single- and multi-cluster Helmfile features cover LLM routing and request handling. See tests/bdd/README.md to run them.

Local Gotchas

  • github.com/olric-data/olric is replaced by github.com/max007-008/olric on branch cas. The fork keeps the upstream module path and adds the CompareAndSwap primitive used by rate limiting.

  • To pick up a new fork commit, run:

    go get github.com/max007-008/olric@cas
go mod tidy
  • If the CAS work is upstreamed, remove the replace directive and bump the upstream requirement.
  • Do not hand-edit generated protobufs or mocks. Regenerate them from the owning source.
  • Chart and stack changes belong in the chart or stack subtree. Keep runtime env/config names aligned when cross-subtree follow-up is needed.

Naming

  • Source subtree: src/invocation-plane-services/llm-api-gateway
  • Go module suffix: llm-gateway
  • Runtime service: llm-api-gateway

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

Check your setup, free