feat(web): DB-backed Jobs tab + scheduler-dispatched local jobs (#436)

* feat(web): DB-backed Jobs tab, scheduler-dispatched local jobs, remove active-jobs-bar

- Remove active-jobs-bar UI element (HTML, CSS, JS polling)
- Move job handlers from server.rs to handlers/jobs.rs
- Remove user_id scoping (single-user gateway)
- Add list_agent_jobs() and agent_job_summary() to Database trait
  (both postgres and libsql backends) for non-sandbox job visibility
- Wire SchedulerSlot into CreateJobTool so execute_local dispatches
  via scheduler (persists to DB + spawns worker) instead of creating
  phantom ContextManager-only jobs
- Update /status and /list slash commands to read from DB for
  consistency with Jobs tab
- Fix worker mark_completed: skip if already terminal or stuck
- Add agent job cancel via DB update in both web handler and slash cmd
- Add Stuck → Completed guard with tracing in worker completion path

Co-Authored-By: Claude Sonnet 4.6 <[email protected]>

* fix: address PR review comments

- Log warning when get_context fails in worker completion path
- Extract duplicated status-counting logic into AgentJobSummary::add_count()
  helper, used by both postgres and libsql backends

Co-Authored-By: Claude Sonnet 4.6 <[email protected]>

---------

Co-authored-by: Claude Sonnet 4.6 <[email protected]>
Co-authored-by: Nick Pismenkov <[email protected]>
This commit is contained in:
Henry Park
2026-02-28 19:58:07 -08:00
committed by GitHub
co-authored by Claude Sonnet 4.6 Nick Pismenkov
parent afb49597ac
commit 6481448d50
14 changed files with 508 additions and 605 deletions
+5
View File
@@ -138,6 +138,11 @@ impl Agent {
// Convenience accessors
/// Get the scheduler (for external wiring, e.g. CreateJobTool).
pub fn scheduler(&self) -> Arc<Scheduler> {
Arc::clone(&self.scheduler)
}
pub(super) fn store(&self) -> Option<&Arc<dyn Database>> {
self.deps.store.as_ref()
}
+89 -7
View File
@@ -12,6 +12,7 @@ use crate::agent::session::Session;
use crate::agent::submission::SubmissionResult;
use crate::agent::{Agent, MessageIntent};
use crate::channels::{IncomingMessage, StatusUpdate};
use crate::context::JobState;
use crate::error::Error;
use crate::llm::{ChatMessage, Reasoning};
@@ -117,6 +118,22 @@ impl Agent {
let uuid = Uuid::parse_str(&id)
.map_err(|_| crate::error::JobError::NotFound { id: Uuid::nil() })?;
// Try DB first for persistent state, fall back to ContextManager.
if let Some(store) = self.store()
&& let Ok(Some(ctx)) = store.get_job(uuid).await
{
return Ok(format!(
"Job: {}\nStatus: {:?}\nCreated: {}\nStarted: {}\nActual cost: {}",
ctx.title,
ctx.state,
ctx.created_at.format("%Y-%m-%d %H:%M:%S"),
ctx.started_at
.map(|t| t.format("%Y-%m-%d %H:%M:%S").to_string())
.unwrap_or_else(|| "Not started".to_string()),
ctx.actual_cost
));
}
let ctx = self.context_manager.get_context(uuid).await?;
if ctx.user_id != user_id {
return Err(crate::error::JobError::NotFound { id: uuid }.into());
@@ -134,10 +151,38 @@ impl Agent {
))
}
None => {
// Show summary of all jobs
// Show summary from DB for consistency with Jobs tab.
if let Some(store) = self.store() {
let mut total = 0;
let mut in_progress = 0;
let mut completed = 0;
let mut failed = 0;
let mut stuck = 0;
if let Ok(s) = store.agent_job_summary().await {
total += s.total;
in_progress += s.in_progress;
completed += s.completed;
failed += s.failed;
stuck += s.stuck;
}
if let Ok(s) = store.sandbox_job_summary().await {
total += s.total;
in_progress += s.running;
completed += s.completed;
failed += s.failed + s.interrupted;
}
return Ok(format!(
"Jobs summary: Total: {} In Progress: {} Completed: {} Failed: {} Stuck: {}",
total, in_progress, completed, failed, stuck
));
}
// Fallback to ContextManager if no DB.
let summary = self.context_manager.summary_for(user_id).await;
Ok(format!(
"Jobs summary:\n Total: {}\n In Progress: {}\n Completed: {}\n Failed: {}\n Stuck: {}",
"Jobs summary: Total: {} In Progress: {} Completed: {} Failed: {} Stuck: {}",
summary.total,
summary.in_progress,
summary.completed,
@@ -159,6 +204,15 @@ impl Agent {
self.scheduler.stop(uuid).await?;
// Also update DB so the Jobs tab reflects cancellation immediately.
if let Some(store) = self.store()
&& let Err(e) = store
.update_job_status(uuid, JobState::Cancelled, Some("Cancelled by user"))
.await
{
tracing::warn!(job_id = %uuid, "Failed to persist cancellation to DB: {}", e);
}
Ok(format!("Job {} has been cancelled.", job_id))
}
@@ -167,21 +221,49 @@ impl Agent {
user_id: &str,
_filter: Option<String>,
) -> Result<String, Error> {
let jobs = self.context_manager.all_jobs_for(user_id).await;
// List from DB for consistency with Jobs tab.
if let Some(store) = self.store() {
let agent_jobs = match store.list_agent_jobs().await {
Ok(jobs) => jobs,
Err(e) => {
tracing::warn!("Failed to list agent jobs: {}", e);
Vec::new()
}
};
let sandbox_jobs = match store.list_sandbox_jobs().await {
Ok(jobs) => jobs,
Err(e) => {
tracing::warn!("Failed to list sandbox jobs: {}", e);
Vec::new()
}
};
if agent_jobs.is_empty() && sandbox_jobs.is_empty() {
return Ok("No jobs found.".to_string());
}
let mut output = String::from("Jobs:\n");
for j in &agent_jobs {
output.push_str(&format!(" {} - {} ({})\n", j.id, j.title, j.status));
}
for j in &sandbox_jobs {
output.push_str(&format!(" {} - {} ({})\n", j.id, j.task, j.status));
}
return Ok(output);
}
// Fallback to ContextManager if no DB.
let jobs = self.context_manager.all_jobs_for(user_id).await;
if jobs.is_empty() {
return Ok("No jobs found.".to_string());
}
let mut output = String::from("Jobs:\n");
for job_id in jobs {
if let Ok(ctx) = self.context_manager.get_context(job_id).await
&& ctx.user_id == user_id
{
if let Ok(ctx) = self.context_manager.get_context(job_id).await {
output.push_str(&format!(" {} - {} ({:?})\n", job_id, ctx.title, ctx.state));
}
}
Ok(output)
}
+31
View File
@@ -158,6 +158,37 @@ Report when the job is complete or if you encounter issues you cannot resolve."#
match result {
Ok(Ok(())) => {
tracing::info!("Worker for job {} completed successfully", self.job_id);
// Only mark completed if still in an active, non-stuck state.
// The execution_loop may have already called mark_completed or
// mark_stuck (e.g. "plan completed but work remains").
let current_state = self
.context_manager()
.get_context(self.job_id)
.await
.map(|ctx| ctx.state);
match current_state {
Ok(state) if state.is_terminal() => {
// Already in a terminal state (e.g. execution_loop
// called mark_completed itself).
}
Ok(JobState::Stuck) => {
// execution_loop marked this as stuck (e.g. "plan
// completed but work remains"); leave for self-repair.
tracing::info!(
"Job {} returned Ok but is Stuck — leaving for self-repair",
self.job_id
);
}
Ok(_) => {
self.mark_completed().await?;
}
Err(e) => {
tracing::warn!(
job_id = %self.job_id,
"Failed to get job context, cannot mark as completed: {}", e
);
}
}
}
Ok(Err(e)) => {
tracing::error!("Worker for job {} failed: {}", self.job_id, e);