Hermes Agent is an open-source AI agent framework developed by Nous Research. Version 0.17.0, codenamed “The Reach Release,” expands communication channels with iMessage and WhatsApp support, enables background sub-agent execution, and introduces image-to-image editing capabilities.
📑Table of Contents
Key New Features in Hermes Agent v0.17.0
Hermes Agent v0.17.0 adds support for multiple new communication platforms. For iMessage, the new Photon Spectrum platform plugin enables direct use without Mac relay or self-hosted solutions like BlueBubbles. Setup involves running hermes photon login with device code authentication.
Official WhatsApp Business Cloud API support simplifies messaging for business use cases. Raft agent network integration allows external agents via a wake-channel bridge, configured with the RAFT_PROFILE environment variable.
The background sub-agent execution feature allows long-running tasks to be processed in the background. Image-to-image editing support lets image generation models learn to edit existing images, enhancing workflows.
Additional improvements include Automation Blueprints enhancements, a major upgrade to the memory tool, a browser overhaul for Skills Hub, and the addition of a dashboard profile builder.
Source: GitHub Release (June 19, 2026)
Release Scale and Background
This release is large-scale with 1,475 commits, approximately 800 merged PRs, and 1,693 files changed. There were 235,390 lines inserted and 50,730 lines deleted, closing over 300 issues. Community contributors number 245.
Positioned as a major update following the v0.16.0 desktop version, the main themes are expanded reach through new communication channels and strengthened background execution. Detailed release notes are available on the official site.
Source: Official GitHub Release Notes, Hermes Agent Official Site
Practical Usage and Impact
In daily conversations, iMessage and WhatsApp can be used directly. For team collaboration, Raft integration enables coordinated multi-agent work. Long-running tasks proceed uninterrupted via background execution, and image editing workflows benefit from the image-to-image feature.
From operational experience, the expanded channels allow seamless invocation of agents within everyday messaging tools. The direct iMessage support is particularly impactful for Mac users.
Comparison and Positioning vs Other Tools
Compared to other AI coding agents like Claude Code or Codex, Hermes Agent stands out for being open-source and its broad platform support. Raft network integration and background sub-agents are practical differentiators not commonly found elsewhere.
Its open-source nature offers high customizability and community-driven extensions.
Frequently Asked Questions (FAQ)
Table: Main New Features Comparison
| Feature | v0.16.0 | v0.17.0 |
|---|---|---|
| iMessage | None (Mac relay) | Photon direct support |
| Background sub-agents | None | Supported |
| Raft | None | Integrated |
| image-to-image | Basic | Edit support |
| Limited | Official Business API support |
Source: GitHub Release v2026.6.19 (as of June 2026)
Related articles:
- Boltz Bio、BoltzMol-1 / BoltzProt-1 をリリース — 創薬向け新モデル、Claude Code / Codex 統合対応
- MiniMax M3 リリース — 初のオープンウェイト frontier モデル(1M コンテキスト + ネイティブ multimodal + 59% SWE-Bench)
- Arbor:Claude Code・Codexを2.5倍上回るHypothesis-Tree AI最適化フレームワーク
Summary
Hermes Agent v0.17.0 significantly improves practicality through expanded communication channels and background execution. iMessage/WhatsApp support and Raft integration facilitate everyday work and team collaboration. Leverage the open-source advantages by checking the official GitHub for the latest information and trying it yourself.
Recommended next step: Clone the GitHub repository, install v0.17.0, and test Photon login.
Source: Official X Announcement, Official Site
Author
krona23
Over 20 years in the IT industry, serving as Division Head and CTO at multiple companies running large-scale web services in Japan. Experienced across Windows, iOS, Android, and web development. Currently focused on AI-native transformation. At DevGENT, sharing practical guides on AI code editors, automation tools, and LLMs in three languages.



![Arbor: Hypothesis-Tree AI Optimization Framework Beats Claude Code & Codex by 2.5x [2026]](https://i0.wp.com/devgent.org/wp-content/uploads/2026/06/aitools-eyecatch-3657.webp?fit=300%2C169&ssl=1)




Leave a Reply