feat(jobs): add admin call to purge jubs result

src/infrastructure/scheduler/pg_job_store.rs

@@ -431,6 +431,30 @@ impl JobStoreProvider for PgJobStoreProvider {
             .collect())
     }
 
+    async fn purge_terminal_runs(&self, retention_days: i32) -> Result<u64, DomainError> {
+        // Defensive floor — zero would eat just-completed runs;
+        // negative would eat the whole terminal history.
+        let days = retention_days.max(1);
+        // `ON DELETE CASCADE` on jobs.run_findings.run_id
+        // (migration 20260930000001) drops findings with their
+        // parent run. Non-terminal statuses (Running / Paused /
+        // CancelRequested) explicitly excluded to protect
+        // in-flight work.
+        let result = sqlx::query(
+            r#"
+            DELETE FROM jobs.recoverable_runs
+             WHERE status IN ('Completed', 'Failed')
+               AND completed_at IS NOT NULL
+               AND completed_at < NOW() - ($1 || ' days')::interval
+            "#,
+        )
+        .bind(days.to_string())
+        .execute(self.pool.as_ref())
+        .await
+        .map_err(|e| map_sqlx_err("purge_terminal_runs", e))?;
+        Ok(result.rows_affected())
+    }
+
     async fn request_cancel(&self, job_name: &str) -> Result<Option<Uuid>, DomainError> {
         // Only Running → CancelRequested flips. `Paused` can be
         // cancelled by not resuming — no need for a state change.

src/infrastructure/scheduler/recoverable.rs

@@ -371,6 +371,29 @@ pub trait JobStoreProvider: Send + Sync {
         &self,
         run_id: Uuid,
     ) -> Result<Vec<(String, u64)>, DomainError>;
+
+    /// Operator-triggered retention cleanup. DELETEs every
+    /// TERMINAL run (`Completed`, `Failed`) whose `completed_at`
+    /// is older than `retention_days` days ago. Findings drop
+    /// alongside via the `ON DELETE CASCADE` FK on
+    /// `jobs.run_findings.run_id`.
+    ///
+    /// Non-terminal rows (`Running`, `Paused`, `CancelRequested`)
+    /// are ALWAYS preserved regardless of age — an in-flight or
+    /// paused run must not be reaped by retention.
+    ///
+    /// `retention_days` is treated as `max(1, retention_days)` at
+    /// the impl layer to defend against a zero/negative value
+    /// eating just-completed runs.
+    ///
+    /// Returns the number of run rows deleted (which equals
+    /// the number of finding rows deleted *transitively* via
+    /// CASCADE; callers wanting the finding count separately
+    /// should query it BEFORE calling this).
+    ///
+    /// Powers `POST /api/admin/jobs/runs/purge`. Not periodic — the
+    /// operator decides when to reclaim space.
+    async fn purge_terminal_runs(&self, retention_days: i32) -> Result<u64, DomainError>;
 }
 
 /// How a `RunProgress` fraction was derived. Lets the UI communicate
@@ -1094,6 +1117,25 @@ mod tests {
             Ok(counts.into_iter().collect())
         }
 
+        async fn purge_terminal_runs(&self, retention_days: i32) -> Result<u64, DomainError> {
+            // Test-double: no `completed_at` to compare against, so
+            // just drop every terminal-state store when
+            // `retention_days` > 0. Sufficient for the trait
+            // contract check; PG impl exercises the real
+            // `completed_at < NOW() - days` filter.
+            let days = retention_days.max(1);
+            if days == 0 {
+                return Ok(0);
+            }
+            let mut stores = self.stores.lock().unwrap();
+            let before = stores.len();
+            stores.retain(|s| {
+                let state = s.state.lock().unwrap();
+                !matches!(state.status, RunStatus::Completed | RunStatus::Failed)
+            });
+            Ok((before - stores.len()) as u64)
+        }
+
         async fn request_cancel(&self, _job_name: &str) -> Result<Option<Uuid>, DomainError> {
             let stores = self.stores.lock().unwrap();
             if let Some(s) = stores.last() {

src/interfaces/api/handlers/admin_handler.rs

@@ -164,6 +164,9 @@ pub fn admin_routes() -> Router<Arc<AppState>> {
             "/jobs/{name}/runs/{id}/findings",
             get(list_job_run_findings),
         )
+        // Retention cleanup — operator-triggered, not periodic.
+        // See `purge_job_runs` docstring for the semantics.
+        .route("/jobs/runs/purge", post(purge_job_runs))
         // Drives — admin-wide view (distinct from `/api/drives` which
         // is filtered to the caller's role grants).
         .route("/drives", get(list_all_drives))
@@ -2392,3 +2395,74 @@ pub async fn list_job_run_findings(
         Err(e) => AppError::internal_error(format!("list_findings failed: {e}")).into_response(),
     }
 }
+
+/// Query parameters for `POST /api/admin/jobs/runs/purge`.
+///
+/// `days` — retention window. Terminal runs (`Completed`, `Failed`)
+/// with `completed_at` older than this many days ago are deleted
+/// (with their findings via CASCADE). Default 30. Minimum enforced
+/// at 1 by the provider — zero would eat runs completed seconds
+/// ago. Non-terminal runs are ALWAYS preserved regardless of age.
+#[derive(serde::Deserialize)]
+pub struct PurgeJobRunsQuery {
+    #[serde(default = "default_purge_days")]
+    pub days: i32,
+}
+
+fn default_purge_days() -> i32 {
+    30
+}
+
+/// `POST /api/admin/jobs/runs/purge?days=N` — operator-triggered
+/// cleanup of old terminal runs + their findings. Not periodic;
+/// admins fire this when they want to reclaim `jobs.*` history
+/// space. Delegates entirely to
+/// `JobStoreProvider::purge_terminal_runs` — no SQL in the handler
+/// (see `AGENTS.md` § handler thinness).
+#[utoipa::path(
+    post,
+    path = "/api/admin/jobs/runs/purge",
+    params(
+        ("days" = Option<i32>, Query, description = "Retention window in days (default 30, minimum 1). Terminal runs older than this are deleted with their findings; non-terminal runs are always preserved."),
+    ),
+    responses(
+        (status = 200, description = "Purge complete"),
+        (status = 401, description = "Unauthorized"),
+        (status = 403, description = "Admin required"),
+        (status = 500, description = "DB error"),
+    ),
+    security(("bearerAuth" = [])),
+    tag = "admin"
+)]
+pub async fn purge_job_runs(
+    State(state): State<Arc<AppState>>,
+    axum::extract::Query(query): axum::extract::Query<PurgeJobRunsQuery>,
+) -> impl IntoResponse {
+    use crate::infrastructure::scheduler::JobStoreProvider as _;
+    let retention_days = query.days.max(1);
+    match state
+        .core
+        .job_store_provider
+        .purge_terminal_runs(retention_days)
+        .await
+    {
+        Ok(purged) => {
+            tracing::info!(
+                target: "audit",
+                event = "jobs.runs_purged",
+                purged = purged,
+                retention_days = retention_days,
+                "👮🏻‍♂️ admin purged {purged} terminal recoverable-run row(s) past {retention_days} day retention (findings cascaded)",
+            );
+            (
+                StatusCode::OK,
+                Json(serde_json::json!({
+                    "purged":         purged,
+                    "retention_days": retention_days,
+                })),
+            )
+                .into_response()
+        }
+        Err(e) => AppError::internal_error(format!("purge failed: {e}")).into_response(),
+    }
+}