<?xml version="1.0" encoding="UTF-8"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom" xmlns:dc="http://purl.org/dc/elements/1.1/">
  <channel>
    <title>DEV Community: Parth Gorde</title>
    <description>The latest articles on DEV Community by Parth Gorde (@wellparth).</description>
    <link>https://dev.to/wellparth</link>
    <image>
      <url>https://media2.dev.to/dynamic/image/width=90,height=90,fit=cover,gravity=auto,format=auto/https:%2F%2Fdev-to-uploads.s3.us-east-2.amazonaws.com%2Fuploads%2Fuser%2Fprofile_image%2F2818356%2Fb9af3fd3-2b91-4fc9-a7af-3fa7b809bcca.jpeg</url>
      <title>DEV Community: Parth Gorde</title>
      <link>https://dev.to/wellparth</link>
    </image>
    <atom:link rel="self" type="application/rss+xml" href="https://dev.to/feed/wellparth"/>
    <language>en</language>
    <item>
      <title>almost-cli — BYOK Multi-Agent Coding, Local First</title>
      <dc:creator>Parth Gorde</dc:creator>
      <pubDate>Wed, 09 Sep 2026 14:48:15 +0000</pubDate>
      <link>https://dev.to/wellparth/almost-cli-byok-multi-agent-coding-local-first-37nf</link>
      <guid>https://dev.to/wellparth/almost-cli-byok-multi-agent-coding-local-first-37nf</guid>
      <description>&lt;p&gt;&lt;strong&gt;Repo:&lt;/strong&gt; &lt;a href="https://github.com/wellparth/almost-cli" rel="noopener noreferrer"&gt;https://github.com/wellparth/almost-cli&lt;/a&gt;&lt;/p&gt;

&lt;p&gt;Most coding assistants rent you a model and keep your code in the cloud. almost-cli takes a different approach: bring your own API keys, run a team of specialized agents locally, and keep everything under your control.&lt;/p&gt;

&lt;h2&gt;
  
  
  The idea
&lt;/h2&gt;

&lt;p&gt;almost-cli is a provider agnostic CLI for multi-agent coding. Connect your own OpenAI, Gemini, DeepSeek, or NVIDIA NIM keys and orchestrate agents: Architect, Planner, Builder, Tester, Debugger against a local codebase. No hosted inference. No backend account. Your keys stay on your machine.&lt;/p&gt;

&lt;p&gt;The core insight is simple: agent identity is separate from model identity. An agent is a configuration of system prompt, tool set, permissions, and workflow. The model is just the runtime. You can route cheap models for planning and strong models for debugging without changing behavior.&lt;/p&gt;

&lt;h2&gt;
  
  
  How it works
&lt;/h2&gt;

&lt;p&gt;The stack is layered for safety and composability:&lt;/p&gt;

&lt;ul&gt;
&lt;li&gt;CLI/TUI talks to an Agent Orchestrator&lt;/li&gt;
&lt;li&gt;Orchestrator coordinates Agent Runtime, Context Engine, and Permission Engine&lt;/li&gt;
&lt;li&gt;Agents run with explicit tool sets: Filesystem, Shell, Git, Search, MCP&lt;/li&gt;
&lt;li&gt;Permissions are deny by default. Every tool call is checked. MCP servers are permission gated per tool.&lt;/li&gt;
&lt;/ul&gt;

&lt;p&gt;This makes local first multi-agent work practical. You get auditability, explicit boundaries, and the ability to swap providers per agent.&lt;/p&gt;

&lt;h2&gt;
  
  
  Why BYOK matters
&lt;/h2&gt;

&lt;p&gt;With BYOK you own your data, your keys, and your agent definitions. There is no vendor lock in and no code leaves your machine unless you ask for it. The trade off is you own ops, which is exactly the point for teams that care about security and control.&lt;/p&gt;

&lt;h2&gt;
  
  
  Repo layout
&lt;/h2&gt;



&lt;div class="highlight js-code-highlight"&gt;
&lt;pre class="highlight codeowners"&gt;&lt;code&gt;&lt;span class="n"&gt;apps/cli&lt;/span&gt;&lt;span class="w"&gt;
&lt;/span&gt;&lt;span class="n"&gt;packages/agent-core&lt;/span&gt;&lt;span class="w"&gt;
&lt;/span&gt;&lt;span class="n"&gt;packages/providers/*&lt;/span&gt;&lt;span class="w"&gt;
&lt;/span&gt;&lt;span class="n"&gt;packages/tools/*&lt;/span&gt;&lt;span class="w"&gt;
&lt;/span&gt;&lt;span class="n"&gt;packages/permissions&lt;/span&gt;&lt;span class="w"&gt;
&lt;/span&gt;&lt;span class="n"&gt;packages/context&lt;/span&gt;&lt;span class="w"&gt;
&lt;/span&gt;&lt;span class="n"&gt;packages/orchestrator&lt;/span&gt;&lt;span class="w"&gt;
&lt;/span&gt;&lt;span class="n"&gt;packages/sessions&lt;/span&gt;&lt;span class="w"&gt;
&lt;/span&gt;&lt;span class="n"&gt;packages/storage&lt;/span&gt;&lt;span class="w"&gt;
&lt;/span&gt;&lt;span class="n"&gt;packages/agent-definitions&lt;/span&gt;&lt;span class="w"&gt;
&lt;/span&gt;&lt;span class="n"&gt;agents/*&lt;/span&gt;&lt;span class="w"&gt;
&lt;/span&gt;&lt;/code&gt;&lt;/pre&gt;

&lt;/div&gt;



&lt;h2&gt;
  
  
  Try it
&lt;/h2&gt;



&lt;div class="highlight js-code-highlight"&gt;
&lt;pre class="highlight shell"&gt;&lt;code&gt;pnpm &lt;span class="nb"&gt;install
&lt;/span&gt;pnpm typecheck
pnpm &lt;span class="nb"&gt;test
&lt;/span&gt;pnpm build
&lt;/code&gt;&lt;/pre&gt;

&lt;/div&gt;



&lt;p&gt;If you build developer tools, agent security, or local first infra, I would love feedback. The repo is open and I am iterating in public.&lt;/p&gt;

&lt;p&gt;Today was day one of building. If you find any issues, please open them. #buildinpublic&lt;/p&gt;

</description>
      <category>ai</category>
      <category>agents</category>
      <category>opensource</category>
      <category>programming</category>
    </item>
  </channel>
</rss>
