llm: switch Skald to agent-loop Model clients; drop llm-client (phase 1, D13)

The LLM call path now runs on the agent-loop crate's clients and trait:

- core-api: BuiltLlmClient.client is Arc<dyn agent_loop::model::Model>;
  chatbot.rs (ChatbotClient + wire types) deleted; APP_NAME re-exported
  from agent-loop
- providers (openai/anthropic/ollama/openrouter/requesty/declared) build
  OpenAiModel/AnthropicModel/OllamaModel with the model's wire id
- LoggingModel decorator (llm/logging.rs) replaces LoggingChatbotClient;
  per-request correlation (session/stack/user) travels in the new
  ModelRequest.log field, never sent to providers
- llm_call/llm_loop/compactor speak Model::complete + ModelResponse;
  retriability via Model::is_retriable (structured status, B6 rule now
  the crate's default); payload persistence reads RawMeta off
  ModelResponse/ModelError
- crates/llm-client and skald-core/src/chatbot deleted

Full workspace test suite green (incl. 162 skald-core + 32 agent-loop).
This commit is contained in:
2026-07-25 23:55:17 +01:00
parent b8cc6d263b
commit 882a8c9cb9
29 changed files with 251 additions and 2128 deletions
+113
View File
@@ -0,0 +1,113 @@
//! Transparent logging decorator for any [`agent_loop::model::Model`].
//!
//! [`LoggingModel`] intercepts every `complete` call, measures the duration,
//! and persists a **metadata-only** row to `llm_requests` in `system.db`
//! (fire-and-forget). Per-request correlation (session/stack/user id) travels
//! in [`ModelRequest::log`], set by the caller; the payload (request/response
//! bodies) is returned to the caller inside [`ModelResponse::raw`] /
//! [`ModelError::raw`] so it can be written to the user's own database.
//!
//! The split keeps conversation content (payloads) behind the user key while
//! metadata (cost, tokens, timing) stays in the admin-readable registry.
//! (Successor of `chatbot::logging::LoggingChatbotClient`, blueprint D13.)
use std::sync::Arc;
use std::time::Instant;
use async_trait::async_trait;
use sqlx::SqlitePool;
use tokio::sync::mpsc;
use tracing::warn;
use agent_loop::model::{Model, ModelError, ModelRequest, ModelResponse, StreamDelta};
use crate::db::llm_requests;
pub struct LoggingModel {
inner: Arc<dyn Model>,
pool: Arc<SqlitePool>,
model_name: String,
}
impl LoggingModel {
pub fn new(inner: Arc<dyn Model>, pool: Arc<SqlitePool>, model_name: impl Into<String>) -> Self {
Self { inner, pool, model_name: model_name.into() }
}
}
#[async_trait]
impl Model for LoggingModel {
async fn complete(
&self,
req: &ModelRequest,
deltas: Option<mpsc::Sender<StreamDelta>>,
) -> Result<ModelResponse, ModelError> {
let start = Instant::now();
let result = self.inner.complete(req, deltas).await;
let duration_ms = start.elapsed().as_millis() as i64;
// Per-request correlation set by the caller (llm_call / compactor).
let log = req.log.clone().unwrap_or_default();
let session_id = log["session_id"].as_i64();
let stack_id = log["stack_id"].as_i64();
let user_id = log["user_id"].as_str().map(str::to_string);
let request_id = Some(req.request_id.clone());
let model_name = self.model_name.clone();
let pool = Arc::clone(&self.pool);
match &result {
Ok(resp) => {
let usage = resp.usage();
let (input_tokens, output_tokens, cache_read, cache_write) = (
usage.input_tokens.map(|n| n as i64),
usage.output_tokens.map(|n| n as i64),
usage.cache_read.map(|n| n as i64),
usage.cache_write.map(|n| n as i64),
);
tokio::spawn(async move {
if let Err(e) = llm_requests::insert(&pool, llm_requests::LlmRequestRow {
request_id,
user_id,
session_id,
stack_id,
model_name,
error_text: None,
input_tokens,
output_tokens,
duration_ms,
cache_read_tokens: cache_read,
cache_creation_tokens: cache_write,
}).await {
warn!(error = %e, "llm_requests: failed to insert log row");
}
});
}
Err(e) => {
let error_text = e.to_string();
tokio::spawn(async move {
if let Err(log_err) = llm_requests::insert(&pool, llm_requests::LlmRequestRow {
request_id,
user_id,
session_id,
stack_id,
model_name,
error_text: Some(error_text),
input_tokens: None,
output_tokens: None,
duration_ms,
cache_read_tokens: None,
cache_creation_tokens: None,
}).await {
warn!(error = %log_err, "llm_requests: failed to insert error log row");
}
});
}
}
result
}
fn is_retriable(&self, err: &ModelError) -> bool {
self.inner.is_retriable(err)
}
}
+4 -4
View File
@@ -8,11 +8,11 @@ use sqlx::SqlitePool;
use tokio::sync::RwLock;
use tracing::{info, warn};
use crate::chatbot::ChatbotClient;
use crate::chatbot::logging::LoggingChatbotClient;
use agent_loop::model::Model;
use core_api::provider::LlmStrength;
use crate::provider::{ApiProvider, ProviderRegistry, ReasoningMode};
use super::logging::LoggingModel;
use super::providers::RemoteLlmModelInfo;
use super::{ClientStatus, LlmEntry, LlmModelInfo, LlmModelRecord, LlmProviderInfo, LlmProviderRecord};
use super::db;
@@ -512,8 +512,8 @@ fn build_entry(
let prompt_cache = built.prompt_cache;
let extra = model.extra_params.clone();
let client: Arc<dyn ChatbotClient> = match log_pool {
Some(pool) => Arc::new(LoggingChatbotClient::new(inner, pool, &model.name)),
let client: Arc<dyn Model> = match log_pool {
Some(pool) => Arc::new(LoggingModel::new(inner, pool, &model.name)),
None => inner,
};
+4 -2
View File
@@ -1,10 +1,12 @@
pub(crate) mod db;
pub mod logging;
pub mod manager;
pub mod providers;
use std::sync::Arc;
use crate::chatbot::ChatbotClient;
use agent_loop::model::Model;
use crate::provider::ServiceType;
pub use core_api::provider::{LlmProviderRecord, LlmModelRecord, LlmStrength, ReasoningMode};
@@ -13,7 +15,7 @@ pub use manager::{LlmManager, sort_models_for_agent};
/// A resolved, ready-to-use LLM client with its associated metadata.
#[derive(Clone)]
pub struct LlmEntry {
pub client: Arc<dyn ChatbotClient>,
pub client: Arc<dyn Model>,
pub model: String,
pub model_db_id: i64,
pub strength: Option<LlmStrength>,
@@ -2,7 +2,7 @@ use std::sync::Arc;
use anyhow::{Context, Result, anyhow};
use crate::chatbot::anthropic::AnthropicClient;
use agent_loop::models::AnthropicModel;
use crate::llm::{LlmModelRecord, LlmProviderRecord};
use crate::llm::providers::{RemoteLlmModelInfo, extra_with_reasoning};
use crate::provider::{ApiProvider, BuiltLlmClient, ProviderField, ProviderUiMeta, ReasoningMode, ServiceType};
@@ -111,7 +111,7 @@ impl ApiProvider for AnthropicProvider {
// stays uncached, as before.
let prompt_cache = model.capabilities.iter().any(|c| c == "tool_search");
Ok(BuiltLlmClient {
client: Arc::new(AnthropicClient::with_extra_body(key, extra)),
client: Arc::new(AnthropicModel::with_extra_body(key, model.model_id.clone(), extra)),
prompt_cache,
})
})())
@@ -18,7 +18,7 @@ use std::sync::Arc;
use anyhow::{anyhow, Context, Result};
use tracing::{info, warn};
use crate::chatbot::openai::OpenAiClient;
use agent_loop::models::OpenAiModel;
use crate::llm::providers::{extra_with_reasoning, RemoteLlmModelInfo};
use crate::llm::{LlmModelRecord, LlmProviderRecord};
use crate::provider::{
@@ -575,7 +575,7 @@ impl ApiProvider for DeclaredProvider {
let extra = extra_with_reasoning(self, model);
let prompt_cache = self.spec.prompt_cache;
Ok(BuiltLlmClient {
client: Arc::new(OpenAiClient::new(self.base_url(record), key, extra, prompt_cache)),
client: Arc::new(OpenAiModel::with_options(self.base_url(record), key, model.model_id.clone(), extra, prompt_cache)),
prompt_cache,
})
})())
+3 -3
View File
@@ -15,7 +15,7 @@ use anyhow::{anyhow, Context, Result};
use core_api::provider::{ApiProvider, BuiltLlmClient, LlmModelRecord, LlmProviderRecord};
use crate::chatbot::openai::OpenAiClient;
use agent_loop::models::OpenAiModel;
/// Computes the `extra_params` an OpenAI-compatible client should be built with,
/// given a model's stored `extra_params` and its selected reasoning value. The
@@ -75,7 +75,7 @@ pub(crate) async fn fetch_openai_models(
.ok_or_else(|| anyhow!("unexpected {who} response shape"))
}
/// Builds an `OpenAiClient` for an OpenAI-compatible provider: requires the
/// Builds an `OpenAiModel` for an OpenAI-compatible provider: requires the
/// provider record's `api_key` and merges the model's stored `extra_params`
/// with the provider-translated reasoning fragment (see `extra_with_reasoning`).
pub(crate) fn build_openai_llm(
@@ -89,7 +89,7 @@ pub(crate) fn build_openai_llm(
.with_context(|| format!("provider '{}': api_key required for {}", record.name, provider.type_id()))?;
let extra = extra_with_reasoning(provider, model);
Ok(BuiltLlmClient {
client: Arc::new(OpenAiClient::new(base_url, key, extra, prompt_cache)),
client: Arc::new(OpenAiModel::with_options(base_url, key, model.model_id.clone(), extra, prompt_cache)),
prompt_cache,
})
}
@@ -2,7 +2,7 @@ use std::sync::Arc;
use anyhow::{Result, anyhow};
use crate::chatbot::ollama::OllamaClient;
use agent_loop::models::OllamaModel;
use crate::llm::{LlmModelRecord, LlmProviderRecord};
use crate::llm::providers::RemoteLlmModelInfo;
use crate::provider::{ApiProvider, BuiltLlmClient, ProviderField, ProviderUiMeta, ServiceType};
@@ -101,9 +101,9 @@ impl ApiProvider for OllamaProvider {
Ok(Some(Self::parse_model_info(&resp, model_id)))
}
fn build_llm(&self, record: &LlmProviderRecord, _model: &LlmModelRecord) -> Option<Result<BuiltLlmClient>> {
fn build_llm(&self, record: &LlmProviderRecord, model: &LlmModelRecord) -> Option<Result<BuiltLlmClient>> {
Some(Ok(BuiltLlmClient {
client: Arc::new(OllamaClient::new(record.base_url.as_deref())),
client: Arc::new(OllamaModel::new(record.base_url.as_deref(), model.model_id.clone())),
prompt_cache: false,
}))
}