vinzenz/hyperhive
vinzenz/hyperhive
Watch
0
0
Fork
You've already forked hyperhive
 0
Code Activity

review classifier: run on ACP when the swarm settings choose it

Browse source 
The swarm settings gain `runtime` (`api-key`, the default, or `acp`) and
an `acp` provider record: an https base URL, the provider's model, and
its prices per million tokens. A PUT now changes only the fields its
body names, so the settings form that sends model, effort and limit
leaves the runtime alone; `null` clears `effort` or `acp`. The body stays
strict, and the stored document keeps #5192's lenient read.

The controller sends the runtime in each run request. On `acp` the
classifier runs one prompt on a new `opencode acp` session per
classification through hive-runtime's AcpRuntime, in a scratch
directory, with a config rendered from the settings. opencode runs
`read` and other tools without asking, so OPENCODE_PERMISSION
{"*":"deny"} takes every tool away from the model, and a deny-all policy
refuses any request on top. The title call, which resends the prompt,
is turned off. The deadline cancels the turn, and a turn that does not
end within 15s of that is dropped, which also covers an agent stuck
before its prompt. The answer is the final assistant text, as ACP has
no result event.

The provider key is read per run from
swarm/services/review-classifier/acp-provider, which the bao UI's
admins login may create and update, and reaches opencode through
AcpRuntime::with_provider_key and the trampoline's allowlist.

Each ACP classification's cost is computed from the configured prices
and the reported tokens, and added to
hyperhive.review_classifier.cost.usage labelled repo, pr, model and
currency.

Part of #5096
This commit is contained in:
atlas 2026-10-13 10:15:27 +02:00 • committed by mara
commit 1b4e764e5c
24 changed files with 1622 additions and 262 deletions
Download patch file Download diff file Expand all files Collapse all files

31
swarm-review-classifier/src/main.rs
View file

@ -7,7 +7,7 @@
//! the controller reads the PR and sends it along.
//!
//! The store identity comes from the `BAO_*` environment. The queue client
//! secret is read at start and again when authelia refuses it, the API key
//! secret is read at start and again when authelia refuses it, the run's key
//! once per run, both from the store under that identity.
use std::path::PathBuf;
@ -19,9 +19,11 @@ use swarm_queue_client::review_classifier::{QUEUE_GROUP, Reply, SUBJECT, TryOut}
use swarm_queue_client::stored_secret::StoredSecret;
use swarm_queue_client::{ClientSecret, QueueConfig};
mod acp;
mod aggregate;
mod answer;
mod child_env;
mod cost;
mod prompt;
mod run;
@ -37,7 +39,8 @@ const PREFIX: &str = "SWARM_REVIEW_CLASSIFIER";
const RETRY_DELAYS_S: [u64; 6] = [1, 2, 4, 8, 16, 30];
fn main() -> Result<()> {
// Before anything else: as the trampoline, this process becomes `claude`.
// Before anything else: as the trampoline, this process becomes the
// model child.
child_env::exec_if_requested();
tokio::runtime::Builder::new_multi_thread()
.enable_all()
@ -54,10 +57,20 @@ async fn serve() -> Result<()> {
std::env::var(&full).with_context(|| format!("{full} is unset"))
};
let client_id = var("OIDC_CLIENT_ID")?;
let claude = run::Claude {
program: PathBuf::from(var("CLAUDE")?),
trampoline: std::env::current_exe()
.context("finding this binary for the claude trampoline")?,
let trampoline =
std::env::current_exe().context("finding this binary for the child trampoline")?;
let models = run::Runtimes {
claude: run::Claude {
program: PathBuf::from(var("CLAUDE")?),
trampoline: trampoline.clone(),
},
acp: acp::Acp {
program: PathBuf::from(var("OPENCODE")?),
trampoline,
deadline: swarm_queue_client::review_classifier::CLASSIFICATION_DEADLINE,
cancel_wait: acp::CANCEL_WAIT,
cost: cost::Otel,
},
};
let secret = client_secret_with_retry(&client_id).await?;
let reread_id = client_id.clone();
@ -81,16 +94,16 @@ async fn serve() -> Result<()> {
.with_context(|| format!("subscribing to {SUBJECT}"))?;
tracing::info!(subject = SUBJECT, "review classifier: awaiting try-outs");
let claude = std::sync::Arc::new(claude);
let models = std::sync::Arc::new(models);
while let Some(msg) = requests.next().await {
let Some(reply_to) = msg.reply.clone() else {
tracing::warn!("a try-out arrived with no reply subject; dropping it");
continue;
};
let queue = queue.clone();
let claude = std::sync::Arc::clone(&claude);
let models = std::sync::Arc::clone(&models);
tokio::spawn(async move {
let reply = handle(&msg.payload, &run::Store, &*claude).await;
let reply = handle(&msg.payload, &run::Store, &*models).await;
match serde_json::to_vec(&reply) {
Ok(body) => {
if let Err(e) = queue.publish(reply_to, body.into()).await {