daarion-admin/microdao-daarion
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
Files
815a28747404aceafc22445c8d0fb6d4cc9a3c33
microdao-daarion/site/tasks/PHASE3_MASTER_TASK/index.html
Apple ef3473db21 snapshot: NODE1 production state 2026-02-09 
Complete snapshot of /opt/microdao-daarion/ from NODE1 (144.76.224.179).
This represents the actual running production code that has diverged
significantly from the previous main branch.

Key changes from old main:
- Gateway (http_api.py): expanded from ~40KB to 164KB with full agent support
- Router: new /v1/agents/{id}/infer endpoint with vision + DeepSeek routing
- Behavior Policy: SOWA v2.2 (3-level: FULL/ACK/SILENT)
- Agent Registry: config/agent_registry.yml as single source of truth
- 13 agents configured (was 3)
- Memory service integration
- CrewAI teams and roles

Excluded from snapshot: venv/, .env, data/, backups, .tgz archives

Co-authored-by: Cursor <cursoragent@cursor.com>
2026-02-09 08:46:46 -08:00

1349 lines
56 KiB
HTML
Raw Blame History

This file contains ambiguous Unicode characters
This file contains Unicode characters that might be confused with other characters. If you think that this is intentional, you can safely ignore this warning. Use the Escape button to reveal them.
<!doctype html>
<html lang="en" class="no-js">
<head>
<meta charset="utf-8">
<meta name="viewport" content="width=device-width,initial-scale=1">
<link rel="canonical" href="https://IvanTytar.github.io/microdao-daarion/tasks/PHASE3_MASTER_TASK/">
<link rel="icon" href="../../assets/images/favicon.png">
<meta name="generator" content="mkdocs-1.5.3, mkdocs-material-9.5.18">
<title>TASK: PHASE 3 — LLM PROXY + MEMORY ORCHESTRATOR + TOOLCORE - DAARION Documentation</title>
<link rel="stylesheet" href="../../assets/stylesheets/main.66ac8b77.min.css">
<link rel="preconnect" href="https://fonts.gstatic.com" crossorigin>
<link rel="stylesheet" href="https://fonts.googleapis.com/css?family=Roboto:300,300i,400,400i,700,700i%7CRoboto+Mono:400,400i,700,700i&display=fallback">
<style>:root{--md-text-font:"Roboto";--md-code-font:"Roboto Mono"}</style>
<script>__md_scope=new URL("../..",location),__md_hash=e=>[...e].reduce((e,_)=>(e<<5)-e+_.charCodeAt(0),0),__md_get=(e,_=localStorage,t=__md_scope)=>JSON.parse(_.getItem(t.pathname+"."+e)),__md_set=(e,_,t=localStorage,a=__md_scope)=>{try{t.setItem(a.pathname+"."+e,JSON.stringify(_))}catch(e){}}</script>
</head>
<body dir="ltr">
<input class="md-toggle" data-md-toggle="drawer" type="checkbox" id="__drawer" autocomplete="off">
<input class="md-toggle" data-md-toggle="search" type="checkbox" id="__search" autocomplete="off">
<label class="md-overlay" for="__drawer"></label>
<div data-md-component="skip">
<a href="#task-phase-3-llm-proxy-memory-orchestrator-toolcore" class="md-skip">
Skip to content
</a>
</div>
<div data-md-component="announce">
</div>
<header class="md-header md-header--shadow" data-md-component="header">
<nav class="md-header__inner md-grid" aria-label="Header">
<a href="../.." title="DAARION Documentation" class="md-header__button md-logo" aria-label="DAARION Documentation" data-md-component="logo">
<svg xmlns="http://www.w3.org/2000/svg" viewBox="0 0 24 24"><path d="M12 8a3 3 0 0 0 3-3 3 3 0 0 0-3-3 3 3 0 0 0-3 3 3 3 0 0 0 3 3m0 3.54C9.64 9.35 6.5 8 3 8v11c3.5 0 6.64 1.35 9 3.54 2.36-2.19 5.5-3.54 9-3.54V8c-3.5 0-6.64 1.35-9 3.54Z"/></svg>
</a>
<label class="md-header__button md-icon" for="__drawer">
<svg xmlns="http://www.w3.org/2000/svg" viewBox="0 0 24 24"><path d="M3 6h18v2H3V6m0 5h18v2H3v-2m0 5h18v2H3v-2Z"/></svg>
</label>
<div class="md-header__title" data-md-component="header-title">
<div class="md-header__ellipsis">
<div class="md-header__topic">
<span class="md-ellipsis">
DAARION Documentation
</span>
</div>
<div class="md-header__topic" data-md-component="header-topic">
<span class="md-ellipsis">
TASK: PHASE 3 — LLM PROXY + MEMORY ORCHESTRATOR + TOOLCORE
</span>
</div>
</div>
</div>
<script>var media,input,key,value,palette=__md_get("__palette");if(palette&&palette.color){"(prefers-color-scheme)"===palette.color.media&&(media=matchMedia("(prefers-color-scheme: light)"),input=document.querySelector(media.matches?"[data-md-color-media='(prefers-color-scheme: light)']":"[data-md-color-media='(prefers-color-scheme: dark)']"),palette.color.media=input.getAttribute("data-md-color-media"),palette.color.scheme=input.getAttribute("data-md-color-scheme"),palette.color.primary=input.getAttribute("data-md-color-primary"),palette.color.accent=input.getAttribute("data-md-color-accent"));for([key,value]of Object.entries(palette.color))document.body.setAttribute("data-md-color-"+key,value)}</script>
<label class="md-header__button md-icon" for="__search">
<svg xmlns="http://www.w3.org/2000/svg" viewBox="0 0 24 24"><path d="M9.5 3A6.5 6.5 0 0 1 16 9.5c0 1.61-.59 3.09-1.56 4.23l.27.27h.79l5 5-1.5 1.5-5-5v-.79l-.27-.27A6.516 6.516 0 0 1 9.5 16 6.5 6.5 0 0 1 3 9.5 6.5 6.5 0 0 1 9.5 3m0 2C7 5 5 7 5 9.5S7 14 9.5 14 14 12 14 9.5 12 5 9.5 5Z"/></svg>
</label>
<div class="md-search" data-md-component="search" role="dialog">
<label class="md-search__overlay" for="__search"></label>
<div class="md-search__inner" role="search">
<form class="md-search__form" name="search">
<input type="text" class="md-search__input" name="query" aria-label="Search" placeholder="Search" autocapitalize="off" autocorrect="off" autocomplete="off" spellcheck="false" data-md-component="search-query" required>
<label class="md-search__icon md-icon" for="__search">
<svg xmlns="http://www.w3.org/2000/svg" viewBox="0 0 24 24"><path d="M9.5 3A6.5 6.5 0 0 1 16 9.5c0 1.61-.59 3.09-1.56 4.23l.27.27h.79l5 5-1.5 1.5-5-5v-.79l-.27-.27A6.516 6.516 0 0 1 9.5 16 6.5 6.5 0 0 1 3 9.5 6.5 6.5 0 0 1 9.5 3m0 2C7 5 5 7 5 9.5S7 14 9.5 14 14 12 14 9.5 12 5 9.5 5Z"/></svg>
<svg xmlns="http://www.w3.org/2000/svg" viewBox="0 0 24 24"><path d="M20 11v2H8l5.5 5.5-1.42 1.42L4.16 12l7.92-7.92L13.5 5.5 8 11h12Z"/></svg>
</label>
<nav class="md-search__options" aria-label="Search">
<button type="reset" class="md-search__icon md-icon" title="Clear" aria-label="Clear" tabindex="-1">
<svg xmlns="http://www.w3.org/2000/svg" viewBox="0 0 24 24"><path d="M19 6.41 17.59 5 12 10.59 6.41 5 5 6.41 10.59 12 5 17.59 6.41 19 12 13.41 17.59 19 19 17.59 13.41 12 19 6.41Z"/></svg>
</button>
</nav>
</form>
<div class="md-search__output">
<div class="md-search__scrollwrap" data-md-scrollfix>
<div class="md-search-result" data-md-component="search-result">
<div class="md-search-result__meta">
Initializing search
</div>
<ol class="md-search-result__list" role="presentation"></ol>
</div>
</div>
</div>
</div>
</div>
</nav>
</header>
<div class="md-container" data-md-component="container">
<main class="md-main" data-md-component="main">
<div class="md-main__inner md-grid">
<div class="md-sidebar md-sidebar--primary" data-md-component="sidebar" data-md-type="navigation" >
<div class="md-sidebar__scrollwrap">
<div class="md-sidebar__inner">
<nav class="md-nav md-nav--primary" aria-label="Navigation" data-md-level="0">
<label class="md-nav__title" for="__drawer">
<a href="../.." title="DAARION Documentation" class="md-nav__button md-logo" aria-label="DAARION Documentation" data-md-component="logo">
<svg xmlns="http://www.w3.org/2000/svg" viewBox="0 0 24 24"><path d="M12 8a3 3 0 0 0 3-3 3 3 0 0 0-3-3 3 3 0 0 0-3 3 3 3 0 0 0 3 3m0 3.54C9.64 9.35 6.5 8 3 8v11c3.5 0 6.64 1.35 9 3.54 2.36-2.19 5.5-3.54 9-3.54V8c-3.5 0-6.64 1.35-9 3.54Z"/></svg>
</a>
DAARION Documentation
</label>
<ul class="md-nav__list" data-md-scrollfix>
<li class="md-nav__item">
<a href="../../public/" class="md-nav__link">
<span class="md-ellipsis">
Home
</span>
</a>
</li>
<li class="md-nav__item">
<a href="../../public/getting-started/" class="md-nav__link">
<span class="md-ellipsis">
Getting Started
</span>
</a>
</li>
<li class="md-nav__item">
<a href="../../public/architecture-overview/" class="md-nav__link">
<span class="md-ellipsis">
Architecture
</span>
</a>
</li>
<li class="md-nav__item">
<a href="../../public/daiS_daos_overview/" class="md-nav__link">
<span class="md-ellipsis">
DAIS & DAOS
</span>
</a>
</li>
<li class="md-nav__item md-nav__item--section md-nav__item--nested">
<input class="md-nav__toggle md-toggle " type="checkbox" id="__nav_5" >
<label class="md-nav__link" for="__nav_5" id="__nav_5_label" tabindex="">
<span class="md-ellipsis">
Internal
</span>
<span class="md-nav__icon md-icon"></span>
</label>
<nav class="md-nav" data-md-level="1" aria-labelledby="__nav_5_label" aria-expanded="false">
<label class="md-nav__title" for="__nav_5">
<span class="md-nav__icon md-icon"></span>
Internal
</label>
<ul class="md-nav__list" data-md-scrollfix>
<li class="md-nav__item md-nav__item--nested">
<input class="md-nav__toggle md-toggle " type="checkbox" id="__nav_5_1" >
<label class="md-nav__link" for="__nav_5_1" id="__nav_5_1_label" tabindex="0">
<span class="md-ellipsis">
Infra
</span>
<span class="md-nav__icon md-icon"></span>
</label>
<nav class="md-nav" data-md-level="2" aria-labelledby="__nav_5_1_label" aria-expanded="false">
<label class="md-nav__title" for="__nav_5_1">
<span class="md-nav__icon md-icon"></span>
Infra
</label>
<ul class="md-nav__list" data-md-scrollfix>
<li class="md-nav__item">
<a href="../../internal/infra/INFRA_AUTOMATION_PACK_V1/" class="md-nav__link">
<span class="md-ellipsis">
Infra Automation Pack v1
</span>
</a>
</li>
<li class="md-nav__item">
<a href="../../internal/infra/monitoring_overview/" class="md-nav__link">
<span class="md-ellipsis">
Monitoring Overview
</span>
</a>
</li>
<li class="md-nav__item">
<a href="../../internal/infra/nodes_registry_v0/" class="md-nav__link">
<span class="md-ellipsis">
Nodes Registry v0
</span>
</a>
</li>
</ul>
</nav>
</li>
<li class="md-nav__item md-nav__item--nested">
<input class="md-nav__toggle md-toggle " type="checkbox" id="__nav_5_2" >
<label class="md-nav__link" for="__nav_5_2" id="__nav_5_2_label" tabindex="0">
<span class="md-ellipsis">
Specs
</span>
<span class="md-nav__icon md-icon"></span>
</label>
<nav class="md-nav" data-md-level="2" aria-labelledby="__nav_5_2_label" aria-expanded="false">
<label class="md-nav__title" for="__nav_5_2">
<span class="md-nav__icon md-icon"></span>
Specs
</label>
<ul class="md-nav__list" data-md-scrollfix>
<li class="md-nav__item">
<a href="../../internal/specs/matrix_presence_aggregator/" class="md-nav__link">
<span class="md-ellipsis">
Matrix Presence Aggregator
</span>
</a>
</li>
<li class="md-nav__item">
<a href="../../internal/specs/city_map_spec/" class="md-nav__link">
<span class="md-ellipsis">
City Map Spec
</span>
</a>
</li>
<li class="md-nav__item">
<a href="../../internal/specs/node_join_protocol_draft/" class="md-nav__link">
<span class="md-ellipsis">
Node Join Protocol (Draft)
</span>
</a>
</li>
</ul>
</nav>
</li>
</ul>
</nav>
</li>
</ul>
</nav>
</div>
</div>
</div>
<div class="md-sidebar md-sidebar--secondary" data-md-component="sidebar" data-md-type="toc" >
<div class="md-sidebar__scrollwrap">
<div class="md-sidebar__inner">
<nav class="md-nav md-nav--secondary" aria-label="Table of contents">
<label class="md-nav__title" for="__toc">
<span class="md-nav__icon md-icon"></span>
Table of contents
</label>
<ul class="md-nav__list" data-md-component="toc" data-md-scrollfix>
<li class="md-nav__item">
<a href="#1-service-llm-proxy-servicesllm-proxy" class="md-nav__link">
<span class="md-ellipsis">
1) SERVICE: LLM PROXY (services/llm-proxy)
</span>
</a>
<nav class="md-nav" aria-label="1) SERVICE: LLM PROXY (services/llm-proxy)">
<ul class="md-nav__list">
<li class="md-nav__item">
<a href="#11-api" class="md-nav__link">
<span class="md-ellipsis">
1.1 API
</span>
</a>
</li>
<li class="md-nav__item">
<a href="#12-model-routing-routerpy" class="md-nav__link">
<span class="md-ellipsis">
1.2 Model routing (router.py)
</span>
</a>
</li>
<li class="md-nav__item">
<a href="#13-providers-providerspy" class="md-nav__link">
<span class="md-ellipsis">
1.3 Providers (providers/*.py)
</span>
</a>
</li>
<li class="md-nav__item">
<a href="#14-logging-limits-middlewarespy-logging_middlewarepy" class="md-nav__link">
<span class="md-ellipsis">
1.4 Logging &amp; limits (middlewares.py, logging_middleware.py)
</span>
</a>
</li>
<li class="md-nav__item">
<a href="#15-readmemd" class="md-nav__link">
<span class="md-ellipsis">
1.5 README.md
</span>
</a>
</li>
</ul>
</nav>
</li>
<li class="md-nav__item">
<a href="#2-service-memory-orchestrator-servicesmemory-orchestrator" class="md-nav__link">
<span class="md-ellipsis">
2) SERVICE: MEMORY ORCHESTRATOR (services/memory-orchestrator)
</span>
</a>
<nav class="md-nav" aria-label="2) SERVICE: MEMORY ORCHESTRATOR (services/memory-orchestrator)">
<ul class="md-nav__list">
<li class="md-nav__item">
<a href="#21-api" class="md-nav__link">
<span class="md-ellipsis">
2.1 API
</span>
</a>
</li>
<li class="md-nav__item">
<a href="#22-backends" class="md-nav__link">
<span class="md-ellipsis">
2.2 Backends
</span>
</a>
</li>
<li class="md-nav__item">
<a href="#23-embedding-client-embedding_clientpy" class="md-nav__link">
<span class="md-ellipsis">
2.3 Embedding client (embedding_client.py)
</span>
</a>
</li>
<li class="md-nav__item">
<a href="#24-integration-points" class="md-nav__link">
<span class="md-ellipsis">
2.4 Integration points
</span>
</a>
</li>
</ul>
</nav>
</li>
<li class="md-nav__item">
<a href="#3-service-toolcore-servicestoolcore" class="md-nav__link">
<span class="md-ellipsis">
3) SERVICE: TOOLCORE (services/toolcore)
</span>
</a>
<nav class="md-nav" aria-label="3) SERVICE: TOOLCORE (services/toolcore)">
<ul class="md-nav__list">
<li class="md-nav__item">
<a href="#31-data-model" class="md-nav__link">
<span class="md-ellipsis">
3.1 Data model
</span>
</a>
</li>
<li class="md-nav__item">
<a href="#32-api" class="md-nav__link">
<span class="md-ellipsis">
3.2 API
</span>
</a>
</li>
<li class="md-nav__item">
<a href="#33-executors" class="md-nav__link">
<span class="md-ellipsis">
3.3 Executors
</span>
</a>
</li>
<li class="md-nav__item">
<a href="#34-registry-registrypy" class="md-nav__link">
<span class="md-ellipsis">
3.4 Registry (registry.py)
</span>
</a>
</li>
<li class="md-nav__item">
<a href="#35-nats-optional-phase-3" class="md-nav__link">
<span class="md-ellipsis">
3.5 NATS (optional для Phase 3)
</span>
</a>
</li>
</ul>
</nav>
</li>
<li class="md-nav__item">
<a href="#4-integration-agent-runtime-llm-proxy-memory-toolcore" class="md-nav__link">
<span class="md-ellipsis">
4) INTEGRATION: agent-runtime + LLM Proxy + Memory + Toolcore
</span>
</a>
<nav class="md-nav" aria-label="4) INTEGRATION: agent-runtime + LLM Proxy + Memory + Toolcore">
<ul class="md-nav__list">
<li class="md-nav__item">
<a href="#41-llm-proxy" class="md-nav__link">
<span class="md-ellipsis">
4.1 Використовувати LLM Proxy
</span>
</a>
</li>
<li class="md-nav__item">
<a href="#42-memory-orchestrator" class="md-nav__link">
<span class="md-ellipsis">
4.2 Використовувати Memory Orchestrator
</span>
</a>
</li>
<li class="md-nav__item">
<a href="#43-tools" class="md-nav__link">
<span class="md-ellipsis">
4.3 Виклик tools
</span>
</a>
</li>
</ul>
</nav>
</li>
<li class="md-nav__item">
<a href="#5-docker-docs" class="md-nav__link">
<span class="md-ellipsis">
5) DOCKER + DOCS
</span>
</a>
<nav class="md-nav" aria-label="5) DOCKER + DOCS">
<ul class="md-nav__list">
<li class="md-nav__item">
<a href="#51-docker-compose" class="md-nav__link">
<span class="md-ellipsis">
5.1 docker-compose
</span>
</a>
</li>
<li class="md-nav__item">
<a href="#52" class="md-nav__link">
<span class="md-ellipsis">
5.2 Документація
</span>
</a>
</li>
</ul>
</nav>
</li>
<li class="md-nav__item">
<a href="#6-acceptance-criteria" class="md-nav__link">
<span class="md-ellipsis">
6) ACCEPTANCE CRITERIA
</span>
</a>
<nav class="md-nav" aria-label="6) ACCEPTANCE CRITERIA">
<ul class="md-nav__list">
<li class="md-nav__item">
<a href="#1-llm-proxy" class="md-nav__link">
<span class="md-ellipsis">
1) LLM Proxy:
</span>
</a>
</li>
<li class="md-nav__item">
<a href="#2-memory-orchestrator" class="md-nav__link">
<span class="md-ellipsis">
2) Memory Orchestrator:
</span>
</a>
</li>
<li class="md-nav__item">
<a href="#3-toolcore" class="md-nav__link">
<span class="md-ellipsis">
3) Toolcore:
</span>
</a>
</li>
<li class="md-nav__item">
<a href="#4-end-to-end" class="md-nav__link">
<span class="md-ellipsis">
4) End-to-end:
</span>
</a>
</li>
</ul>
</nav>
</li>
<li class="md-nav__item">
<a href="#end-of-task" class="md-nav__link">
<span class="md-ellipsis">
END OF TASK
</span>
</a>
</li>
</ul>
</nav>
</div>
</div>
</div>
<div class="md-content" data-md-component="content">
<article class="md-content__inner md-typeset">
<h1 id="task-phase-3-llm-proxy-memory-orchestrator-toolcore">TASK: PHASE 3 — LLM PROXY + MEMORY ORCHESTRATOR + TOOLCORE<a class="headerlink" href="#task-phase-3-llm-proxy-memory-orchestrator-toolcore" title="Permanent link">&para;</a></h1>
<p><strong>Goal:</strong>
Зробити агентів DAARION по-справжньому розумними та інструментальними:
- єдиний LLM Proxy з підтримкою кількох провайдерів і локальних моделей;
- Memory Orchestrator, який дає єдиний API для памʼяті агентів;
- Toolcore — реєстр інструментів і безпечне виконання tools для агентів.</p>
<p><strong>PHASE 3 = інфраструктура мислення й дії для Agent Runtime.</strong></p>
<p><strong>Existing (assume from Phase 1–2):</strong>
- messaging-service, matrix-gateway, Messenger UI.
- agent-filter, DAGI Router (router.invoke.agent).
- agent-runtime (використовує /internal/llm/proxy та /internal/agent-memory/*, але поки можуть бути stubs).
- базова БД (users, microdaos, agents, channels, messages, etc.).
- NATS JetStream, PostgreSQL, docker-compose.</p>
<p><strong>Deliverables:</strong>
1. services/llm-proxy/ — LLM Router + Providers + Logging.
2. services/memory-orchestrator/ — єдиний API для памʼяті агентів.
3. services/toolcore/ — registry + execution engine для tools.
4. Інтеграція з agent-runtime.
5. Оновлення docker-compose та docs.</p>
<hr />
<h2 id="1-service-llm-proxy-servicesllm-proxy">1) SERVICE: LLM PROXY (services/llm-proxy)<a class="headerlink" href="#1-service-llm-proxy-servicesllm-proxy" title="Permanent link">&para;</a></h2>
<p><strong>Create folder:</strong></p>
<div class="codehilite"><pre><span></span><code>services/llm-proxy/
main.py
models.py
router.py
providers/
__init__.py
openai_provider.py
deepseek_provider.py
local_provider.py
config.yaml
middlewares.py
logging_middleware.py
Dockerfile
requirements.txt
README.md
</code></pre></div>
<h3 id="11-api">1.1 API<a class="headerlink" href="#11-api" title="Permanent link">&para;</a></h3>
<p><strong>Base URL (internal-only):</strong>
- <code>POST /internal/llm/proxy</code></p>
<p><strong>Request (v1):</strong></p>
<div class="codehilite"><pre><span></span><code><span class="p">{</span>
<span class="w"> </span><span class="nt">&quot;model&quot;</span><span class="p">:</span><span class="w"> </span><span class="s2">&quot;gpt-4.1-mini&quot;</span><span class="p">,</span>
<span class="w"> </span><span class="nt">&quot;messages&quot;</span><span class="p">:</span><span class="w"> </span><span class="p">[</span>
<span class="w"> </span><span class="p">{</span><span class="w"> </span><span class="nt">&quot;role&quot;</span><span class="p">:</span><span class="w"> </span><span class="s2">&quot;system&quot;</span><span class="p">,</span><span class="w"> </span><span class="nt">&quot;content&quot;</span><span class="p">:</span><span class="w"> </span><span class="s2">&quot;...&quot;</span><span class="w"> </span><span class="p">},</span>
<span class="w"> </span><span class="p">{</span><span class="w"> </span><span class="nt">&quot;role&quot;</span><span class="p">:</span><span class="w"> </span><span class="s2">&quot;user&quot;</span><span class="p">,</span><span class="w"> </span><span class="nt">&quot;content&quot;</span><span class="p">:</span><span class="w"> </span><span class="s2">&quot;...&quot;</span><span class="w"> </span><span class="p">}</span>
<span class="w"> </span><span class="p">],</span>
<span class="w"> </span><span class="nt">&quot;metadata&quot;</span><span class="p">:</span><span class="w"> </span><span class="p">{</span>
<span class="w"> </span><span class="nt">&quot;agent_id&quot;</span><span class="p">:</span><span class="w"> </span><span class="s2">&quot;agent:sofia&quot;</span><span class="p">,</span>
<span class="w"> </span><span class="nt">&quot;microdao_id&quot;</span><span class="p">:</span><span class="w"> </span><span class="s2">&quot;microdao:7&quot;</span><span class="p">,</span>
<span class="w"> </span><span class="nt">&quot;channel_id&quot;</span><span class="p">:</span><span class="w"> </span><span class="s2">&quot;...&quot;</span>
<span class="w"> </span><span class="p">}</span>
<span class="p">}</span>
</code></pre></div>
<p><strong>Response:</strong></p>
<div class="codehilite"><pre><span></span><code><span class="p">{</span>
<span class="w"> </span><span class="nt">&quot;content&quot;</span><span class="p">:</span><span class="w"> </span><span class="s2">&quot;текст відповіді&quot;</span><span class="p">,</span>
<span class="w"> </span><span class="nt">&quot;usage&quot;</span><span class="p">:</span><span class="w"> </span><span class="p">{</span>
<span class="w"> </span><span class="nt">&quot;prompt_tokens&quot;</span><span class="p">:</span><span class="w"> </span><span class="mi">123</span><span class="p">,</span>
<span class="w"> </span><span class="nt">&quot;completion_tokens&quot;</span><span class="p">:</span><span class="w"> </span><span class="mi">45</span><span class="p">,</span>
<span class="w"> </span><span class="nt">&quot;total_tokens&quot;</span><span class="p">:</span><span class="w"> </span><span class="mi">168</span>
<span class="w"> </span><span class="p">},</span>
<span class="w"> </span><span class="nt">&quot;provider&quot;</span><span class="p">:</span><span class="w"> </span><span class="s2">&quot;openai&quot;</span><span class="p">,</span>
<span class="w"> </span><span class="nt">&quot;model_resolved&quot;</span><span class="p">:</span><span class="w"> </span><span class="s2">&quot;gpt-4.1-mini&quot;</span>
<span class="p">}</span>
</code></pre></div>
<p><strong>Optional (stub for later):</strong>
- <code>/internal/llm/batch</code>
- <code>/internal/llm/stream</code> (можна залишити TODO)</p>
<h3 id="12-model-routing-routerpy">1.2 Model routing (router.py)<a class="headerlink" href="#12-model-routing-routerpy" title="Permanent link">&para;</a></h3>
<p>Створити модуль:</p>
<div class="codehilite"><pre><span></span><code><span class="k">def</span><span class="w"> </span><span class="nf">route_model</span><span class="p">(</span><span class="n">logical_model</span><span class="p">:</span> <span class="nb">str</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">ProviderConfig</span><span class="p">:</span>
<span class="c1"># Читає зі структури в config.yaml</span>
<span class="c1"># Напр.:</span>
<span class="c1"># - gpt-4.1-mini → OpenAI (gpt-4.1-mini)</span>
<span class="c1"># - deepseek-r1 → DeepSeek API</span>
<span class="c1"># - dagi-local-8b → Local provider (Ollama / vllm)</span>
</code></pre></div>
<p><strong>config.yaml (приклад):</strong></p>
<div class="codehilite"><pre><span></span><code><span class="nt">providers</span><span class="p">:</span>
<span class="w"> </span><span class="nt">openai</span><span class="p">:</span>
<span class="w"> </span><span class="nt">base_url</span><span class="p">:</span><span class="w"> </span><span class="s">&quot;https://api.openai.com/v1&quot;</span>
<span class="w"> </span><span class="nt">api_key_env</span><span class="p">:</span><span class="w"> </span><span class="s">&quot;OPENAI_API_KEY&quot;</span>
<span class="w"> </span><span class="nt">deepseek</span><span class="p">:</span>
<span class="w"> </span><span class="nt">base_url</span><span class="p">:</span><span class="w"> </span><span class="s">&quot;https://api.deepseek.com/v1&quot;</span>
<span class="w"> </span><span class="nt">api_key_env</span><span class="p">:</span><span class="w"> </span><span class="s">&quot;DEEPSEEK_API_KEY&quot;</span>
<span class="w"> </span><span class="nt">local</span><span class="p">:</span>
<span class="w"> </span><span class="nt">base_url</span><span class="p">:</span><span class="w"> </span><span class="s">&quot;http://local-llm:8000&quot;</span>
<span class="nt">models</span><span class="p">:</span>
<span class="w"> </span><span class="nt">gpt-4.1-mini</span><span class="p">:</span>
<span class="w"> </span><span class="nt">provider</span><span class="p">:</span><span class="w"> </span><span class="s">&quot;openai&quot;</span>
<span class="w"> </span><span class="nt">name</span><span class="p">:</span><span class="w"> </span><span class="s">&quot;gpt-4.1-mini&quot;</span>
<span class="w"> </span><span class="nt">deepseek-r1</span><span class="p">:</span>
<span class="w"> </span><span class="nt">provider</span><span class="p">:</span><span class="w"> </span><span class="s">&quot;deepseek&quot;</span>
<span class="w"> </span><span class="nt">name</span><span class="p">:</span><span class="w"> </span><span class="s">&quot;deepseek-r1&quot;</span>
<span class="w"> </span><span class="nt">dagi-local-8b</span><span class="p">:</span>
<span class="w"> </span><span class="nt">provider</span><span class="p">:</span><span class="w"> </span><span class="s">&quot;local&quot;</span>
<span class="w"> </span><span class="nt">name</span><span class="p">:</span><span class="w"> </span><span class="s">&quot;qwen2.5-8b-instruct&quot;</span>
</code></pre></div>
<h3 id="13-providers-providerspy">1.3 Providers (providers/*.py)<a class="headerlink" href="#13-providers-providerspy" title="Permanent link">&para;</a></h3>
<p>Кожен провайдер:</p>
<div class="codehilite"><pre><span></span><code><span class="k">class</span><span class="w"> </span><span class="nc">BaseProvider</span><span class="p">(</span><span class="n">Protocol</span><span class="p">):</span>
<span class="k">async</span> <span class="k">def</span><span class="w"> </span><span class="nf">chat</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">messages</span><span class="p">:</span> <span class="nb">list</span><span class="p">[</span><span class="n">ChatMessage</span><span class="p">],</span> <span class="n">model_name</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">LLMResponse</span><span class="p">:</span> <span class="o">...</span>
</code></pre></div>
<p><strong>Файли:</strong>
- <code>providers/openai_provider.py</code>
- <code>providers/deepseek_provider.py</code>
- <code>providers/local_provider.py</code></p>
<p>Вони:
- отримують normalized messages,
- викликають відповідний API/endpoint,
- повертають LLMResponse (контент + usage).</p>
<p><strong>Local provider:</strong>
- для Phase 3 можна зробити простий HTTP-запит до існуючого локального сервісу (Ollama/vLLM), або stub.</p>
<h3 id="14-logging-limits-middlewarespy-logging_middlewarepy">1.4 Logging &amp; limits (middlewares.py, logging_middleware.py)<a class="headerlink" href="#14-logging-limits-middlewarespy-logging_middlewarepy" title="Permanent link">&para;</a></h3>
<p>Добавити:
- простий rate limit per-agent (стабово, через in-memory або Redis stub).
- логування викликів:
- agent_id, microdao_id, model, latency, tokens.
- TODO hooks для Billing/Usage Engine (на майбутнє).</p>
<h3 id="15-readmemd">1.5 README.md<a class="headerlink" href="#15-readmemd" title="Permanent link">&para;</a></h3>
<p>Описати:
- підтримувані моделі та мапінг,
- як додати нового провайдера,
- як викликати /internal/llm/proxy з інших сервісів (особливо agent-runtime).</p>
<hr />
<h2 id="2-service-memory-orchestrator-servicesmemory-orchestrator">2) SERVICE: MEMORY ORCHESTRATOR (services/memory-orchestrator)<a class="headerlink" href="#2-service-memory-orchestrator-servicesmemory-orchestrator" title="Permanent link">&para;</a></h2>
<p><strong>Goal:</strong>
Єдиний API для роботи з памʼяттю:
- short-term (канальний контекст / event log),
- mid-term (agent memory / RAG embedding store),
- long-term (knowledge base, docs, roadmaps).</p>
<p><strong>Create folder:</strong></p>
<div class="codehilite"><pre><span></span><code>services/memory-orchestrator/
main.py
models.py
router.py
backends/
__init__.py
short_term_pg.py
vector_store_pg.py
kb_filesystem.py
embedding_client.py
config.yaml
Dockerfile
requirements.txt
README.md
</code></pre></div>
<h3 id="21-api">2.1 API<a class="headerlink" href="#21-api" title="Permanent link">&para;</a></h3>
<p><strong>Base URL:</strong>
- <code>POST /internal/agent-memory/query</code>
- <code>POST /internal/agent-memory/store</code>
- <code>POST /internal/agent-memory/summarize</code> (optional v1)</p>
<p><strong>Request /query:</strong></p>
<div class="codehilite"><pre><span></span><code><span class="p">{</span>
<span class="w"> </span><span class="nt">&quot;agent_id&quot;</span><span class="p">:</span><span class="w"> </span><span class="s2">&quot;agent:sofia&quot;</span><span class="p">,</span>
<span class="w"> </span><span class="nt">&quot;microdao_id&quot;</span><span class="p">:</span><span class="w"> </span><span class="s2">&quot;microdao:7&quot;</span><span class="p">,</span>
<span class="w"> </span><span class="nt">&quot;channel_id&quot;</span><span class="p">:</span><span class="w"> </span><span class="s2">&quot;optional&quot;</span><span class="p">,</span>
<span class="w"> </span><span class="nt">&quot;query&quot;</span><span class="p">:</span><span class="w"> </span><span class="s2">&quot;коротко, які були останні зміни в цьому microDAO?&quot;</span><span class="p">,</span>
<span class="w"> </span><span class="nt">&quot;limit&quot;</span><span class="p">:</span><span class="w"> </span><span class="mi">5</span>
<span class="p">}</span>
</code></pre></div>
<p><strong>Response:</strong></p>
<div class="codehilite"><pre><span></span><code><span class="p">{</span>
<span class="w"> </span><span class="nt">&quot;items&quot;</span><span class="p">:</span><span class="w"> </span><span class="p">[</span>
<span class="w"> </span><span class="p">{</span>
<span class="w"> </span><span class="nt">&quot;id&quot;</span><span class="p">:</span><span class="w"> </span><span class="s2">&quot;mem-123&quot;</span><span class="p">,</span>
<span class="w"> </span><span class="nt">&quot;kind&quot;</span><span class="p">:</span><span class="w"> </span><span class="s2">&quot;conversation | kb | task | dao-event&quot;</span><span class="p">,</span>
<span class="w"> </span><span class="nt">&quot;score&quot;</span><span class="p">:</span><span class="w"> </span><span class="mf">0.87</span><span class="p">,</span>
<span class="w"> </span><span class="nt">&quot;content&quot;</span><span class="p">:</span><span class="w"> </span><span class="s2">&quot;Текст фрагменту памʼяті&quot;</span><span class="p">,</span>
<span class="w"> </span><span class="nt">&quot;meta&quot;</span><span class="p">:</span><span class="w"> </span><span class="p">{</span>
<span class="w"> </span><span class="nt">&quot;source&quot;</span><span class="p">:</span><span class="w"> </span><span class="s2">&quot;channel:...&quot;</span><span class="p">,</span>
<span class="w"> </span><span class="nt">&quot;timestamp&quot;</span><span class="p">:</span><span class="w"> </span><span class="s2">&quot;...&quot;</span>
<span class="w"> </span><span class="p">}</span>
<span class="w"> </span><span class="p">}</span>
<span class="w"> </span><span class="p">]</span>
<span class="p">}</span>
</code></pre></div>
<p><strong>Request /store:</strong></p>
<div class="codehilite"><pre><span></span><code><span class="p">{</span>
<span class="w"> </span><span class="nt">&quot;agent_id&quot;</span><span class="p">:</span><span class="w"> </span><span class="s2">&quot;agent:sofia&quot;</span><span class="p">,</span>
<span class="w"> </span><span class="nt">&quot;microdao_id&quot;</span><span class="p">:</span><span class="w"> </span><span class="s2">&quot;microdao:7&quot;</span><span class="p">,</span>
<span class="w"> </span><span class="nt">&quot;channel_id&quot;</span><span class="p">:</span><span class="w"> </span><span class="s2">&quot;optional&quot;</span><span class="p">,</span>
<span class="w"> </span><span class="nt">&quot;kind&quot;</span><span class="p">:</span><span class="w"> </span><span class="s2">&quot;conversation | kb | task | dao-event&quot;</span><span class="p">,</span>
<span class="w"> </span><span class="nt">&quot;content&quot;</span><span class="p">:</span><span class="w"> </span><span class="p">{</span>
<span class="w"> </span><span class="nt">&quot;user_message&quot;</span><span class="p">:</span><span class="w"> </span><span class="s2">&quot;...&quot;</span><span class="p">,</span>
<span class="w"> </span><span class="nt">&quot;agent_reply&quot;</span><span class="p">:</span><span class="w"> </span><span class="s2">&quot;...&quot;</span>
<span class="w"> </span><span class="p">}</span>
<span class="p">}</span>
</code></pre></div>
<p><strong>Response:</strong></p>
<div class="codehilite"><pre><span></span><code><span class="p">{</span><span class="w"> </span><span class="nt">&quot;ok&quot;</span><span class="p">:</span><span class="w"> </span><span class="kc">true</span><span class="p">,</span><span class="w"> </span><span class="nt">&quot;id&quot;</span><span class="p">:</span><span class="w"> </span><span class="s2">&quot;mem-123&quot;</span><span class="w"> </span><span class="p">}</span>
</code></pre></div>
<h3 id="22-backends">2.2 Backends<a class="headerlink" href="#22-backends" title="Permanent link">&para;</a></h3>
<p><strong>backends/short_term_pg.py:</strong>
- зберігає сирі events (або посилання на messages) в PostgreSQL, якщо треба окремо від messenger.</p>
<p><strong>backends/vector_store_pg.py:</strong>
- використовує embeddings (через embedding_client.py) і зберігає в таблиці:
<code>sql
agent_memories (id, agent_id, microdao_id, embedding vector, content, meta)</code>
- простий cosine similarity пошук.</p>
<p><strong>backends/kb_filesystem.py:</strong>
- stub для long-term знань (roadmaps, docs), можна залишити TODO.</p>
<h3 id="23-embedding-client-embedding_clientpy">2.3 Embedding client (embedding_client.py)<a class="headerlink" href="#23-embedding-client-embedding_clientpy" title="Permanent link">&para;</a></h3>
<ul>
<li>Простий клієнт до існуючої embedding-моделі (наприклад, BGE-M3 чи інша) через /internal/embedding/proxy або безпосередньо.</li>
<li>Якщо embedding-сервіс не описаний — зробити stub + TODO.</li>
</ul>
<h3 id="24-integration-points">2.4 Integration points<a class="headerlink" href="#24-integration-points" title="Permanent link">&para;</a></h3>
<p><strong>agent-runtime:</strong>
- замінити прямі виклики agent-memory на:
- <code>POST /internal/agent-memory/query</code>
- <code>POST /internal/agent-memory/store</code></p>
<p>Надалі можна буде підʼєднати:
- microDAO knowledge,
- DAO events,
- roadmaps.</p>
<hr />
<h2 id="3-service-toolcore-servicestoolcore">3) SERVICE: TOOLCORE (services/toolcore)<a class="headerlink" href="#3-service-toolcore-servicestoolcore" title="Permanent link">&para;</a></h2>
<p><strong>Goal:</strong>
Єдина точка для:
- реєстрації tools (інструментів агентів),
- перевірки дозволів,
- виконання tools та повернення результату агенту.</p>
<p><strong>Create folder:</strong></p>
<div class="codehilite"><pre><span></span><code><span class="n">services</span><span class="o">/</span><span class="n">toolcore</span><span class="o">/</span>
<span class="w"> </span><span class="n">main</span><span class="o">.</span><span class="n">py</span>
<span class="w"> </span><span class="n">models</span><span class="o">.</span><span class="n">py</span>
<span class="w"> </span><span class="n">router</span><span class="o">.</span><span class="n">py</span>
<span class="w"> </span><span class="n">registry</span><span class="o">.</span><span class="n">py</span>
<span class="w"> </span><span class="n">executors</span><span class="o">/</span>
<span class="w"> </span><span class="n">__init__</span><span class="o">.</span><span class="n">py</span>
<span class="w"> </span><span class="n">http_executor</span><span class="o">.</span><span class="n">py</span>
<span class="w"> </span><span class="n">python_executor</span><span class="o">.</span><span class="n">py</span><span class="w"> </span><span class="p">(</span><span class="n">optional</span><span class="p">)</span>
<span class="w"> </span><span class="n">config</span><span class="o">.</span><span class="n">yaml</span>
<span class="w"> </span><span class="n">Dockerfile</span>
<span class="w"> </span><span class="n">requirements</span><span class="o">.</span><span class="n">txt</span>
<span class="w"> </span><span class="n">README</span><span class="o">.</span><span class="n">md</span>
</code></pre></div>
<h3 id="31-data-model">3.1 Data model<a class="headerlink" href="#31-data-model" title="Permanent link">&para;</a></h3>
<p><strong>models.py:</strong></p>
<div class="codehilite"><pre><span></span><code><span class="k">class</span><span class="w"> </span><span class="nc">ToolDefinition</span><span class="p">(</span><span class="n">BaseModel</span><span class="p">):</span>
<span class="nb">id</span><span class="p">:</span> <span class="nb">str</span> <span class="c1"># &quot;projects.list&quot;</span>
<span class="n">name</span><span class="p">:</span> <span class="nb">str</span>
<span class="n">description</span><span class="p">:</span> <span class="nb">str</span>
<span class="n">input_schema</span><span class="p">:</span> <span class="nb">dict</span> <span class="c1"># JSON Schema</span>
<span class="n">output_schema</span><span class="p">:</span> <span class="nb">dict</span> <span class="c1"># JSON Schema</span>
<span class="n">executor</span><span class="p">:</span> <span class="n">Literal</span><span class="p">[</span><span class="s2">&quot;http&quot;</span><span class="p">,</span> <span class="s2">&quot;python&quot;</span><span class="p">]</span>
<span class="n">target</span><span class="p">:</span> <span class="nb">str</span> <span class="c1"># HTTP URL or python path</span>
<span class="n">allowed_agents</span><span class="p">:</span> <span class="nb">list</span><span class="p">[</span><span class="nb">str</span><span class="p">]</span> <span class="o">|</span> <span class="kc">None</span>
<span class="k">class</span><span class="w"> </span><span class="nc">ToolCallRequest</span><span class="p">(</span><span class="n">BaseModel</span><span class="p">):</span>
<span class="n">tool_id</span><span class="p">:</span> <span class="nb">str</span>
<span class="n">agent_id</span><span class="p">:</span> <span class="nb">str</span>
<span class="n">microdao_id</span><span class="p">:</span> <span class="nb">str</span>
<span class="n">args</span><span class="p">:</span> <span class="nb">dict</span>
<span class="n">context</span><span class="p">:</span> <span class="nb">dict</span> <span class="o">|</span> <span class="kc">None</span> <span class="o">=</span> <span class="kc">None</span>
<span class="k">class</span><span class="w"> </span><span class="nc">ToolCallResult</span><span class="p">(</span><span class="n">BaseModel</span><span class="p">):</span>
<span class="n">ok</span><span class="p">:</span> <span class="nb">bool</span>
<span class="n">result</span><span class="p">:</span> <span class="nb">dict</span> <span class="o">|</span> <span class="kc">None</span> <span class="o">=</span> <span class="kc">None</span>
<span class="n">error</span><span class="p">:</span> <span class="nb">str</span> <span class="o">|</span> <span class="kc">None</span> <span class="o">=</span> <span class="kc">None</span>
</code></pre></div>
<h3 id="32-api">3.2 API<a class="headerlink" href="#32-api" title="Permanent link">&para;</a></h3>
<p><strong>HTTP (internal):</strong></p>
<ul>
<li>
<p><code>GET /internal/tools</code>
→ список доступних tools</p>
</li>
<li>
<p><code>POST /internal/tools/call</code>
Body: ToolCallRequest
Behavior:</p>
</li>
<li>перевірити, чи agent_id має право на tool_id (allowed_agents).</li>
<li>знайти ToolDefinition.</li>
<li>виконати через відповідний executor.</li>
<li>повернути ToolCallResult.</li>
</ul>
<h3 id="33-executors">3.3 Executors<a class="headerlink" href="#33-executors" title="Permanent link">&para;</a></h3>
<p><strong>executors/http_executor.py:</strong>
- викликає target як HTTP endpoint (POST з args + context).
- хендлінг помилок та таймаутів.</p>
<p><strong>executors/python_executor.py (optional):</strong>
- allows direct call of internal python functions (у v1 можна залишити TODO або мінімальний stub).</p>
<h3 id="34-registry-registrypy">3.4 Registry (registry.py)<a class="headerlink" href="#34-registry-registrypy" title="Permanent link">&para;</a></h3>
<p>Початок — з config.yaml (static registry):</p>
<p><strong>config.yaml:</strong></p>
<div class="codehilite"><pre><span></span><code><span class="nt">tools</span><span class="p">:</span>
<span class="w"> </span><span class="p p-Indicator">-</span><span class="w"> </span><span class="nt">id</span><span class="p">:</span><span class="w"> </span><span class="s">&quot;projects.list&quot;</span>
<span class="w"> </span><span class="nt">name</span><span class="p">:</span><span class="w"> </span><span class="s">&quot;List</span><span class="nv"> </span><span class="s">projects&quot;</span>
<span class="w"> </span><span class="nt">description</span><span class="p">:</span><span class="w"> </span><span class="s">&quot;Повертає</span><span class="nv"> </span><span class="s">список</span><span class="nv"> </span><span class="s">проєктів</span><span class="nv"> </span><span class="s">microDAO.&quot;</span>
<span class="w"> </span><span class="nt">input_schema</span><span class="p">:</span>
<span class="w"> </span><span class="nt">type</span><span class="p">:</span><span class="w"> </span><span class="s">&quot;object&quot;</span>
<span class="w"> </span><span class="nt">properties</span><span class="p">:</span>
<span class="w"> </span><span class="nt">microdao_id</span><span class="p">:</span><span class="w"> </span><span class="p p-Indicator">{</span><span class="nt"> type</span><span class="p">:</span><span class="w"> </span><span class="s">&quot;string&quot;</span><span class="w"> </span><span class="p p-Indicator">}</span>
<span class="w"> </span><span class="nt">required</span><span class="p">:</span><span class="w"> </span><span class="p p-Indicator">[</span><span class="s">&quot;microdao_id&quot;</span><span class="p p-Indicator">]</span>
<span class="w"> </span><span class="nt">output_schema</span><span class="p">:</span>
<span class="w"> </span><span class="nt">type</span><span class="p">:</span><span class="w"> </span><span class="s">&quot;array&quot;</span>
<span class="w"> </span><span class="nt">items</span><span class="p">:</span><span class="w"> </span><span class="p p-Indicator">{</span><span class="nt"> type</span><span class="p">:</span><span class="w"> </span><span class="s">&quot;object&quot;</span><span class="w"> </span><span class="p p-Indicator">}</span>
<span class="w"> </span><span class="nt">executor</span><span class="p">:</span><span class="w"> </span><span class="s">&quot;http&quot;</span>
<span class="w"> </span><span class="nt">target</span><span class="p">:</span><span class="w"> </span><span class="s">&quot;http://projects-service:8000/internal/tools/projects.list&quot;</span>
<span class="w"> </span><span class="nt">allowed_agents</span><span class="p">:</span><span class="w"> </span><span class="p p-Indicator">[</span><span class="s">&quot;agent:sofia&quot;</span><span class="p p-Indicator">,</span><span class="w"> </span><span class="s">&quot;agent:pm&quot;</span><span class="p p-Indicator">,</span><span class="w"> </span><span class="s">&quot;agent:cto&quot;</span><span class="p p-Indicator">]</span>
</code></pre></div>
<p><strong>Registry API:</strong>
- load from config.yaml at startup
- later можна зробити DB-backed registry.</p>
<h3 id="35-nats-optional-phase-3">3.5 NATS (optional для Phase 3)<a class="headerlink" href="#35-nats-optional-phase-3" title="Permanent link">&para;</a></h3>
<p>Можна додати:
- subject: <code>tool.request</code>
- subject: <code>tool.response</code></p>
<p>Але для Phase 3 достатньо HTTP-контракту, який викликає agent-runtime.</p>
<hr />
<h2 id="4-integration-agent-runtime-llm-proxy-memory-toolcore">4) INTEGRATION: agent-runtime + LLM Proxy + Memory + Toolcore<a class="headerlink" href="#4-integration-agent-runtime-llm-proxy-memory-toolcore" title="Permanent link">&para;</a></h2>
<p><strong>Update agent-runtime:</strong></p>
<h3 id="41-llm-proxy">4.1 Використовувати LLM Proxy<a class="headerlink" href="#41-llm-proxy" title="Permanent link">&para;</a></h3>
<p>Замість прямого виклику будь-яких LLM:
- завжди викликати:
<code>http
POST /internal/llm/proxy
{
"model": "&lt;logical model from agent blueprint&gt;",
"messages": [...],
"metadata": {
"agent_id": ...,
"microdao_id": ...,
"channel_id": ...
}
}</code></p>
<h3 id="42-memory-orchestrator">4.2 Використовувати Memory Orchestrator<a class="headerlink" href="#42-memory-orchestrator" title="Permanent link">&para;</a></h3>
<p>При channel_message:
- контекст:
<code>GET /internal/messaging/channels/{channelId}/messages?limit=50</code>
- памʼять:
<code>POST /internal/agent-memory/query</code>
- запис:
<code>POST /internal/agent-memory/store</code> після відповіді.</p>
<h3 id="43-tools">4.3 Виклик tools<a class="headerlink" href="#43-tools" title="Permanent link">&para;</a></h3>
<p>Якщо в blueprint агента передбачені tools:
- парсити вихід LLM (JSON/structured) або через simple convention (наприклад, спеціальний формат "TOOL: projects.list {...}").
- для кожного tool call:
<code>http
POST /internal/tools/call
{
"tool_id": "...",
"agent_id": "...",
"microdao_id": "...",
"args": {...}
}</code>
- включати результати tools у наступний LLM виклик як контекст.
- (На Phase 3 можна реалізувати один-два прості tools: projects.list, followups.create.)</p>
<hr />
<h2 id="5-docker-docs">5) DOCKER + DOCS<a class="headerlink" href="#5-docker-docs" title="Permanent link">&para;</a></h2>
<h3 id="51-docker-compose">5.1 docker-compose<a class="headerlink" href="#51-docker-compose" title="Permanent link">&para;</a></h3>
<p>Додати new services:
- llm-proxy
- memory-orchestrator
- toolcore</p>
<p>Переконатися, що:
- вони в тій же мережі, що agent-runtime, messaging-service, NATS, DB;
- є healthcheck-и (GET /health).</p>
<h3 id="52">5.2 Документація<a class="headerlink" href="#52" title="Permanent link">&para;</a></h3>
<p>Додати/оновити:
- <code>docs/LLM_PROXY_SPEC.md</code>
- <code>docs/MEMORY_ORCHESTRATOR_SPEC.md</code>
- <code>docs/TOOLCORE_SPEC.md</code>
- оновити <code>docs/INDEX.md</code> (додати посилання на ці три сервіси).
- в PHASE2_READY.md або окремому PHASE3_READY.md описати нову архітектуру agent pipeline.</p>
<hr />
<h2 id="6-acceptance-criteria">6) ACCEPTANCE CRITERIA<a class="headerlink" href="#6-acceptance-criteria" title="Permanent link">&para;</a></h2>
<h3 id="1-llm-proxy">1) LLM Proxy:<a class="headerlink" href="#1-llm-proxy" title="Permanent link">&para;</a></h3>
<ul>
<li>✅ /internal/llm/proxy приймає запити, повертає відповіді.</li>
<li>✅ Принаймні 2 логічні моделі працюють (наприклад, один remote, один local.stub).</li>
<li>✅ Логуються виклики per-agent.</li>
</ul>
<h3 id="2-memory-orchestrator">2) Memory Orchestrator:<a class="headerlink" href="#2-memory-orchestrator" title="Permanent link">&para;</a></h3>
<ul>
<li>✅ /internal/agent-memory/query повертає релевантні фрагменти (можна з простим vector search або stub).</li>
<li>✅ /internal/agent-memory/store зберігає нові entries.</li>
<li>✅ agent-runtime успішно використовує його при відповіді.</li>
</ul>
<h3 id="3-toolcore">3) Toolcore:<a class="headerlink" href="#3-toolcore" title="Permanent link">&para;</a></h3>
<ul>
<li>✅ /internal/tools повертає список інструментів.</li>
<li>✅ /internal/tools/call виконує хоча б один реальний tool (HTTP executor).</li>
<li>✅ agent-runtime може викликати хоча б один tool впродовж свого pipeline.</li>
</ul>
<h3 id="4-end-to-end">4) End-to-end:<a class="headerlink" href="#4-end-to-end" title="Permanent link">&para;</a></h3>
<ul>
<li>✅ Агент у Messenger:</li>
<li>читає контекст (messages + memory),</li>
<li>викликає LLM через LLM Proxy,</li>
<li>за потреби викликає tool через Toolcore,</li>
<li>повертає відповідь у канал.</li>
<li>✅ Для одного демо-агента (наприклад, Sofia-Prime) цей ланцюг стабільно працює.</li>
</ul>
<hr />
<h2 id="end-of-task">END OF TASK<a class="headerlink" href="#end-of-task" title="Permanent link">&para;</a></h2>
<p><strong>Estimated Time:</strong> 6-8 weeks<br />
<strong>Priority:</strong> High<br />
<strong>Dependencies:</strong> Phase 2 complete<br />
<strong>Version:</strong> 1.0.0<br />
<strong>Date:</strong> 2025-11-24</p>
</article>
</div>
<script>var target=document.getElementById(location.hash.slice(1));target&&target.name&&(target.checked=target.name.startsWith("__tabbed_"))</script>
</div>
</main>
<footer class="md-footer">
<div class="md-footer-meta md-typeset">
<div class="md-footer-meta__inner md-grid">
<div class="md-copyright">
Made with
<a href="https://squidfunk.github.io/mkdocs-material/" target="_blank" rel="noopener">
Material for MkDocs
</a>
</div>
</div>
</div>
</footer>
</div>
<div class="md-dialog" data-md-component="dialog">
<div class="md-dialog__inner md-typeset"></div>
</div>
<script id="__config" type="application/json">{"base": "../..", "features": ["navigation.sections", "navigation.instant", "content.code.copy"], "search": "../../assets/javascripts/workers/search.b8dbb3d2.min.js", "translations": {"clipboard.copied": "Copied to clipboard", "clipboard.copy": "Copy to clipboard", "search.result.more.one": "1 more on this page", "search.result.more.other": "# more on this page", "search.result.none": "No matching documents", "search.result.one": "1 matching document", "search.result.other": "# matching documents", "search.result.placeholder": "Type to start searching", "search.result.term.missing": "Missing", "select.version": "Select version"}}</script>
<script src="../../assets/javascripts/bundle.3220b9d7.min.js"></script>
</body>
</html>
Reference in New Issue View Git Blame Copy Permalink