Hermes Agent v0.9.0 adds web dashboard, iMessage support, and Fast Mode
TL;DR
Nous Research released Hermes Agent v0.9.0 on April 13. The Everywhere Release adds a browser-based local dashboard, Fast Mode priority queuing for OpenAI and Anthropic models, iMessage and WeChat support (bringing the platform count to 16), background process watchers, and the deepest security hardening pass in the project's history.
What changed
What shipped
Hermes Agent v0.9.0, codenamed The Everywhere Release, landed April 13, 2026. Five capability drops:
Web dashboard
Hermes now has a local browser-based UI. Configure settings, monitor sessions, browse installed skills, and manage the gateway without editing config files. For anyone who has lived in the YAML so far, this is the first time Hermes feels like a proper desktop app.
Fast Mode
New /fast command enables priority processing for OpenAI (GPT-5.4, Codex) and Anthropic (Claude) models via provider priority queues. Latency drops significantly on both providers when Fast Mode is on. Use it when you want the agent to feel instant rather than contemplative.
16 messaging platforms
Hermes now reaches users through 16 messaging channels. v0.9.0 adds four:
- iMessage via BlueBubbles (macOS bridge).
- WeChat / Weixin for the Chinese ecosystem.
- WeCom callback mode for enterprise Chinese deployments.
- Termux / Android native support.
The existing Telegram, Discord, Slack, WhatsApp, Signal, Matrix, Teams, and the rest all continue to work.
Background process monitoring
A new watch_patterns feature notifies the agent in real time when a watched string appears in a running process: errors, "listening on port X", build log signals, any regex you care about. No more "tail this log and tell me when X happens" polling scripts.
Security hardening
v0.9.0 is the deepest security pass in Hermes's history. Highlights:
- •Path traversal protection across the tool system.
- •Shell injection guards on all exec surfaces.
- •SSRF redirect guards when the agent follows URLs.
- •Twilio webhook signature validation (fixes a prior SMS RCE).
- •API server auth enforcement (previously optional in some modes).
Availability
Update via the standard Hermes channel. No breaking changes called out; existing profiles and config carry forward.
Who this matters for
- Vibe Builder: Web dashboard removes the YAML barrier. iMessage integration means Hermes reaches you wherever you are. Fast Mode when you want the agent to feel snappy.
- Basic User: The browser-based dashboard is the first truly approachable way to configure Hermes. Add iMessage to reach it from your phone without a new app.
- Developer: Background process monitoring replaces every log-tail polling script you have written. Security hardening (shell injection, SSRF, path traversal) means self-hosted deployments are production-safer than ever.
What to watch next
The Everywhere Release earns its codename. v0.9.0 is the first release where Hermes stops being a CLI tool with messaging integrations and starts being a full platform you can operate through whichever interface makes sense for the moment: terminal for coding, iMessage for quick questions on the move, web dashboard for monitoring and configuration.
The web dashboard is the sleeper feature. Hermes has always been powerful but intimidating for anyone not comfortable in YAML. A browser UI for settings, sessions, skills, and gateway management lowers the on-ramp dramatically. Vibe builders who bounced off the terminal version in February now have a way in.
Fast Mode is the feature I did not expect. Priority queues on OpenAI and Anthropic mean Hermes can compete with commercial agents on latency, not just capability. The catch: provider priority isn't free, so expect this to be a paid-tier feature or to come with bill-watching.
iMessage via BlueBubbles is the coverage move that matters most in the US market. Being reachable on iMessage, not just Telegram, removes the "why would I chat with my AI on a new app" friction that has capped adoption for every chat-based agent so far.
The security hardening pass is the least glamorous and most important. Self-hosted agents with shell access are a threat model most users underestimate. v0.9.0 closing shell injection, SSRF, and path traversal is what separates "cool project" from "you can actually run this against your real systems." That maturity curve is what distinguishes Hermes from every other open-source agent shipping right now.
by Harsh Desai
About hermes-agent
View the full hermes-agent page →All hermes-agent updatesMore from hermes-agent
- FeatureHermes Agent v0.4.0 adds OpenAI-compatible API, 6 messaging platforms, and @file context
Hermes Agent v0.4.0 (The Platform Expansion Release) shipped March 24, 2026. It exposes Hermes as a drop-in OpenAI-compatible /v1/chat/completions endpoint, adds 6 new messaging adapters (Signal, DingTalk, SMS via Twilio, Mattermost, Matrix, Webhook) bringing the count to 9, introduces @file and @url context injection, adds 4 new inference providers, and enables gateway prompt caching by default.
- FeatureHermes Agent v0.5.0 expands to 400+ models and adds supply-chain hardening
Hermes Agent v0.5.0 (The Hardening Release) shipped March 28, 2026. Nous Portal now routes to 400+ models through one endpoint. Hugging Face becomes a first-class inference provider. Telegram gets private chat topics with per-topic skill binding. Plugin lifecycle hooks activate. Supply-chain hardening removes a compromised dep, pins version ranges, and adds CI scanning.
- FeatureHermes Agent v0.6.0 lands profiles, MCP server mode, and official Docker container
Hermes Agent v0.6.0 (The Multi-Instance Release) shipped March 30, 2026. Profiles let you run multiple isolated Hermes instances from one install. MCP Server Mode exposes the agent's conversations and sessions to Claude Desktop, Cursor, VS Code, and any MCP client. Official Docker container, two new messaging platforms (Feishu/Lark, WeCom), and an ordered fallback provider chain round out the release.