actor trait improvements and ui implementation

core/supervisor/docs/protocol.md

@@ -45,13 +45,27 @@ Jobs can have dependencies on other jobs, which are stored in the `dependencies`
 
 ### Work Queues
 
-Jobs are queued for execution using Redis lists:
+Jobs are queued for execution using Redis lists with the following naming convention:
 ```
-hero:work_queue:{actor_id}
+hero:job:actor_queue:{script_type_suffix}
 ```
 
+Where `{script_type_suffix}` corresponds to the script type:
+- `osis` for OSIS actors (Rhai/HeroScript execution)
+- `sal` for SAL actors (System Abstraction Layer)
+- `v` for V actors (V language execution)
+- `python` for Python actors
+
+**Examples:**
+- OSIS actor queue: `hero:job:actor_queue:osis`
+- SAL actor queue: `hero:job:actor_queue:sal`
+- V actor queue: `hero:job:actor_queue:v`
+- Python actor queue: `hero:job:actor_queue:python`
+
 Actors listen on their specific queue using `BLPOP` for job IDs to process.
 
+**Important:** Actors must use the same queue naming convention in their `actor_id()` method to ensure proper job dispatch. The actor should return `"actor_queue:{script_type_suffix}"` as its actor ID.
+
 ### Stop Queues
 
 Job stop requests are sent through dedicated stop queues:
@@ -63,12 +77,26 @@ Actors monitor these queues to receive stop requests for running jobs.
 
 ### Reply Queues
 
-For synchronous job execution, dedicated reply queues are used:
-```
-hero:reply:{job_id}
-```
+Reply queues are used for responses to specific requests:
 
-Actors send results to these queues when jobs complete.
+- `hero:reply:{request_id}`: Response to a specific request
+
+### Result and Error Queues
+
+When actors process jobs, they store results and errors in two places:
+
+1. **Job Hash Storage**: Results are stored in the job hash fields:
+   - `hero:job:{job_id}` hash with `output` field for results
+   - `hero:job:{job_id}` hash with `error` field for errors
+
+2. **Dedicated Queues**: Results and errors are also pushed to dedicated queues for asynchronous retrieval:
+   - `hero:job:{job_id}:result`: Queue containing job result (use `LPOP` to retrieve)
+   - `hero:job:{job_id}:error`: Queue containing job error (use `LPOP` to retrieve)
+
+This dual storage approach allows clients to:
+- Access results/errors directly from job hash for immediate retrieval
+- Listen on result/error queues for asynchronous notification of job completion
+- Use `BLPOP` on result/error queues for blocking waits on job completion
 
 ## Job Lifecycle