agents: run cron Scheduler inside the long-running daemon (HIR-119)

CHANGELOG.md

@@ -18,6 +18,7 @@ The format is loosely based on [Keep a Changelog](https://keepachangelog.com/en/
 - `examples/cron_executor.rs` — end-to-end example: schedules a webhook to fire on the next per-second tick and exits within ~1 second.
 - README section: **Action Executors → Cron**.
 - `cron` and `chrono` dependencies (chrono with `clock` + `serde` features).
+- `src/scheduler_loop.rs` — long-running tokio task spawned by `pypes start` that loads every cron action across all agents on boot, fires due entries through the existing webhook executor, advances the scheduler, and logs each fire to the daemon's stderr (`~/.agents/tmp/daemon.err`). `POST /agents` now triggers an in-process reload so newly stored cron actions go live without restarting the daemon.
 
 ### Changed
 - `Agent.actions` storage is unchanged on disk (`Vec<String>`), but each entry is now interpreted as a typed `Action` enum at execution time. Strings that don't parse fall through as `Unrecognized` rather than blocking the pipeline.

Cargo.toml

@@ -27,3 +27,4 @@ chrono = { version = "0.4", default-features = false, features = ["clock", "serd
 
 [dev-dependencies]
 wiremock = "0.5"
+tempfile = "3"

README.md

@@ -204,16 +204,25 @@ The `expression` field accepts:
 - **7-field cron with year** — passed through to the underlying parser.
 
 `pypes agent <NAME> run` reports the next computed fire time for each cron
-action without firing it; the actual firing happens via the scheduler:
+action without firing it; the actual firing happens automatically once the
+server is running:
 
 ```bash
 pypes agent scheduled-pinger run
 # [0] cron `*/5 * * * *` → next fire 2026-05-03 12:05:00 UTC
 ```
 
+`pypes start` (daemonized or `--attatch`) launches a background scheduler
+loop that loads every cron action across all agents on boot, sleeps until
+the next due tick, fires the wrapped action (currently `webhook`), and
+advances. Each fire is logged to the daemon stderr at
+`~/.agents/tmp/daemon.err`. Creating a new agent via `POST /agents`
+triggers an in-process reload so newly stored cron actions become live
+without restarting the daemon.
+
 Out of scope for v1: distributed scheduling, persistent missed-fire catchup
-across restarts. The scheduler is single-process and fires only while it's
-running.
+across daemon restarts (the loop starts fresh from `Utc::now()` on boot),
+and per-tenant isolation.
 
 ### Worked example
 

src/lib.rs

@@ -3,4 +3,5 @@ pub mod daemon;
 pub mod db;
 pub mod executors;
 pub mod resource;
+pub mod scheduler_loop;
 pub mod server;