feat(psf): complete high-integrity loop for advanced features
This commit is contained in:
80
notes/org-skill-sub-agent-manager.org
Normal file
80
notes/org-skill-sub-agent-manager.org
Normal file
@@ -0,0 +1,80 @@
|
||||
#+TITLE: SKILL: Sub-Agent Manager (Concurrency & Parallelism)
|
||||
#+ID: skill-sub-agent-manager
|
||||
#+STARTUP: content
|
||||
#+FILETAGS: :concurrency:parallelism:threads:psf:
|
||||
|
||||
* Overview
|
||||
The **Sub-Agent Manager** enables the Neurosymbolic Lisp Machine to handle multiple concurrent thoughts. It allows the primary kernel to "spawn" lightweight, isolated Lisp threads (sub-agents) to perform long-running or background tasks (research, massive refactors, etc.) without blocking the main event bus.
|
||||
|
||||
* Phase A: Demand (PRD)
|
||||
:PROPERTIES:
|
||||
:STATUS: FROZEN
|
||||
:END:
|
||||
|
||||
** 1. Purpose
|
||||
Define the interfaces for parallel cognitive execution and thread lifecycle management.
|
||||
|
||||
** 2. User Needs
|
||||
- **Non-Blocking Execution:** Spawn background threads for long-running tasks.
|
||||
- **Context Isolation:** Sub-agents must have their own execution context to prevent parent context poisoning.
|
||||
- **Communication Loop:** Sub-agents must inject a "Return Stimulus" upon completion.
|
||||
- **Observability:** Ability to list and terminate active sub-agents.
|
||||
|
||||
** 3. Success Criteria
|
||||
*** TODO Successful spawning of a non-blocking background thread
|
||||
*** TODO Verification of context isolation (distinct local variables)
|
||||
*** TODO Autonomous injection of :sub-agent-complete stimulus
|
||||
*** TODO Thread safety verification using bordeaux-threads locks
|
||||
|
||||
* Phase B: Blueprint (PROTOCOL)
|
||||
:PROPERTIES:
|
||||
:STATUS: SIGNED
|
||||
:END:
|
||||
|
||||
** 1. Architectural Intent
|
||||
Interfaces for parallel cognitive loops. Source of truth is the OS thread registry and the kernel event bus.
|
||||
|
||||
** 2. Semantic Interfaces
|
||||
#+begin_src lisp
|
||||
(defun sub-agent-spawn (goal context)
|
||||
"Creates a new thread and starts a localized cognitive loop.")
|
||||
|
||||
(defun sub-agent-list-active ()
|
||||
"Returns a list of currently running sub-agent threads.")
|
||||
#+end_src
|
||||
|
||||
* Phase D: Build (Implementation)
|
||||
|
||||
** Parallel Spawning
|
||||
#+begin_src lisp :tangle projects/org-skill-sub-agent-manager/src/concurrency-logic.lisp
|
||||
(defvar *active-sub-agents* '() "Registry of active sub-agent thread objects.")
|
||||
|
||||
(defun sub-agent-spawn (goal parent-context)
|
||||
(let ((thread-name (format nil "sub-agent-~a" (get-universal-time))))
|
||||
(kernel-log "CONCURRENCY - Spawning sub-agent for goal: ~a" goal)
|
||||
(let ((new-thread
|
||||
(bt:make-thread
|
||||
(lambda ()
|
||||
(handler-case
|
||||
(let* ((context `(:type :SUB-GOAL :payload (:goal ,goal :parent ,parent-context)))
|
||||
(result (org-agent:think context))) ; Execute sub-goal thinking
|
||||
;; Inject the result back into the main kernel bus
|
||||
(org-agent:inject-stimulus
|
||||
`(:type :EVENT :payload (:sensor :sub-agent-complete :result ,result :goal ,goal))))
|
||||
(error (c)
|
||||
(kernel-log "SUB-AGENT ERROR (~a): ~a" thread-name c))))
|
||||
:name thread-name)))
|
||||
(push new-thread *active-sub-agents*)
|
||||
(format nil "SUCCESS - Sub-agent '~a' is now thinking in the background." thread-name))))
|
||||
#+end_src
|
||||
|
||||
* Registration
|
||||
#+begin_src lisp
|
||||
(defskill :skill-sub-agent-manager
|
||||
:priority 90
|
||||
:trigger (lambda (context) (eq (getf (getf context :payload) :action) :spawn))
|
||||
:neuro (lambda (context) nil)
|
||||
:symbolic (lambda (action context)
|
||||
(let ((goal (getf (getf action :payload) :goal)))
|
||||
(sub-agent-spawn goal context))))
|
||||
#+end_src
|
||||
Reference in New Issue
Block a user