49 lines
1.8 KiB
Org Mode
49 lines
1.8 KiB
Org Mode
#+TITLE: SKILL: The Economist Agent (Universal Literate Note)
|
|
#+ID: skill-economist
|
|
#+STARTUP: content
|
|
#+FILETAGS: :economics:optimization:budget:psf:
|
|
#+DEPENDS_ON: skill-router skill-performance-auditor
|
|
|
|
* Overview
|
|
The **Economist Agent** manages the PSF's compute resources. It predicts the "Cost of Thought" for a task and autonomously selects the most cost-effective LLM provider or local model based on complexity and remaining budget.
|
|
|
|
* Phase A: Demand (PRD)
|
|
:PROPERTIES:
|
|
:STATUS: FROZEN
|
|
:END:
|
|
|
|
** 1. Purpose
|
|
Optimize token usage and compute overhead without sacrificing architectural integrity.
|
|
|
|
** 2. User Needs
|
|
- **Predictive Budgeting:** Estimate token cost before triggering SOTA models.
|
|
- **Provider Switching:** Dynamically route tasks between local (Ollama) and cloud (Gemini) models.
|
|
- **Audit Reports:** Provide transparency on compute consumption.
|
|
|
|
* Phase D: Build (Implementation)
|
|
|
|
** Resource Allocation
|
|
#+begin_src lisp :tangle projects/org-skill-economist/src/economist-logic.lisp
|
|
(defun economist-route-task (complexity)
|
|
"Selects the optimal model backend based on task complexity (1-10)."
|
|
(let ((budget (get-current-token-budget)))
|
|
(cond
|
|
((> complexity 8) :gemini-1.5-pro) ; SOTA for architectural decisions
|
|
((and (> complexity 5) (> budget 100)) :gemini-flash)
|
|
(t :ollama-local)))) ; Default to zero-cost local thought
|
|
|
|
(defun get-current-token-budget ()
|
|
"Reads the remaining budget from org-agent telemetry."
|
|
;; Placeholder for actual telemetry lookup
|
|
1000)
|
|
#+end_src
|
|
|
|
* Registration
|
|
#+begin_src lisp
|
|
(defskill :skill-economist
|
|
:priority 95
|
|
:trigger (lambda (context) (eq (getf (getf context :payload) :sensor) :budget-audit))
|
|
:neuro (lambda (context) "Analyze current compute efficiency and propose routing updates.")
|
|
:symbolic (lambda (action context) action))
|
|
#+end_src
|