Hermes Agent Changelog
Release notes, new features, and updates tracked daily. Sourced from the official Hermes Agent changelog at github.com.
- Featurevv0.14.0
Enable LSP semantic diagnostics on file writes
Run a real language server against edited files during `write_file` or `patch` operations to surface type errors, undefined symbols, and missing imports directly to the agent.
Source ↗ - App Updatevv0.14.0
Reduce cold-start launch times by up to 19 seconds
Shave ~19 seconds off the `hermes` launch time by deferring heavy imports, caching model catalogs on disk, parallelizing doctor checks, and skipping the welcome banner in single-query mode.
Source ↗ - App Updatevv0.14.0
Publish Hermes Agent as a native PyPI package
Hermes Agent is now available directly on PyPI. Users can install the agent and its full TUI experience using `pip install hermes-agent` without needing to clone the repository.
Source ↗ - App Updatevv0.14.0
Optimize installation size with lazy-loading dependencies
Heavy backends like messaging adapters, image-gen SDKs, and voice/TTS providers are now lazy-installed on first use. This reduces the initial installation footprint and speeds up setup.
Source ↗ - Featurevv0.14.0
Add native X (Twitter) search tool with OAuth support
Introduce the `x_search` tool, allowing the agent to search the X timeline, find threads, and surface posts directly from the chat using either X OAuth login or an API key.
Source ↗ - App Updatevv0.14.0
Launch native Windows support in early beta
Hermes now runs natively on Windows via cmd.exe and PowerShell without requiring WSL. Includes a dedicated PowerShell installer and fixes for path normalization and process management.
Source ↗ - Integrationvv0.14.0
Add native support for LINE and SimpleX Chat
Hermes expands its messaging reach to 22 platforms with the addition of LINE and the privacy-focused SimpleX Chat. Both are implemented as first-class messaging adapters.
Source ↗ - App Updatevv0.14.0
Optimize browser console evaluations by 180x
Browser tool calls now utilize a persistent CDP connection to Chrome instead of re-initializing sessions. This reduces interaction latency from seconds to milliseconds.
Source ↗ - Featurevv0.14.0
Enable cross-session Claude prompt caching
System prompts, skills, and memory now cache for one hour across sessions when using Claude. This reduces latency and costs for new sessions and background memory reviews.
Source ↗ - App Updatevv0.14.0
Implement lazy-loading and debloated installation
Installations are now significantly lighter as heavy backends for messaging and media SDKs only install on first use. The `[all]` extra has been trimmed to reduce disk footprint and vulnerabilities.
Source ↗ - Integrationvv0.14.0
Deploy end-to-end Microsoft Teams integration
Hermes now features full Microsoft Teams support via Graph auth and webhook listeners. The agent can read messages and post outbound delivery to channels, DMs, and group chats.
Source ↗ - Featurevv0.14.0
Introduce OpenAI-compatible local proxy for OAuth providers
The new `hermes proxy` command creates a local endpoint that translates OpenAI API calls to authenticated OAuth providers like Claude Pro and ChatGPT Pro. This allows tools like Aider and Cline to use existing subscriptions.
Source ↗ - Model Releasevv0.14.0
Launch xAI Grok via SuperGrok OAuth with 1M context
Users can now authenticate with xAI accounts to use Grok-4.3 without separate API keys. The integration supports a 1M token context window and includes an SSH-to-tunnel flow for remote OAuth completion.
Source ↗ - App Updatevv2026.4.23
Hermes Agent adds Bedrock support and a new interface (v0.11.0)
Hermes Agent v0.11.0 ships a full Ink-based TUI rewrite, a pluggable transport layer with native AWS Bedrock support, five new inference paths, QQBot as a 17th messaging platform, and GPT-5.5 access via Codex OAuth.
Source ↗ - Feature
Nous Research releases Hermes Agent v0.10.0 with 118 skills, learning loop
Hermes Agent v0.10.0 introduces a four-layer learning loop for skills, memory, context, and user models. You get 118 bundled skills and 200+ model support for self-improving agents.
Source ↗ - Featurev0.10.0
Launch Nous Tool Gateway
Paid Nous Portal subscribers now have integrated access to web search, image generation, text-to-speech, and browser automation without needing separate API keys. Tools are enabled via the hermes config and managed through the existing subscription.
Source ↗ - Feature
Hermes Agent v0.9.0 Adds Web Dashboard, iMessage Support
Hermes Agent v0.9.0 provides browser dashboard and Fast Mode queuing for OpenAI models. Connect to 16 platforms including iMessage.
Source ↗ - Featurev0.9.0
Comprehensive Security Hardening
Applied a major security update including path traversal protection, shell injection neutralization, SSRF guards for image uploads, and webhook signature validation across all platforms.
Source ↗ - Featurev0.9.0
Add Configuration Backup and Restore
New hermes backup and hermes import commands allow for full snapshots of agent configurations, sessions, skills, and memory, facilitating easy migration between machines.
Source ↗ - Featurev0.9.0
Introduce Pluggable Context Engine
Context management is now modular via hermes plugins. Users can swap in custom engines to control agent visibility, filtering, summarization, or domain-specific context injection.
Source ↗ - Featurev0.9.0
Add xAI and Xiaomi MiMo Providers
First-class support for xAI (Grok) and Xiaomi MiMo providers, including direct API access, model catalog integration, and setup wizard support.
Source ↗ - Featurev0.9.0
Implement Background Process Monitoring
Use watch_patterns to monitor background process output in real-time. Receive notifications when specific events occur, such as error logs or service readiness, without manual polling.
Source ↗ - Featurev0.9.0
Support Native Android via Termux
Run Hermes Agent natively on Android devices using Termux. Includes optimized TUI for mobile screens, voice backend support, and on-device image generation capabilities.
Source ↗ - Integrationv0.9.0
Add iMessage and WeChat Support
Hermes now supports iMessage via BlueBubbles and native WeChat/WeCom integration, bringing the agent to major messaging ecosystems with full support for media uploads and streaming.
Source ↗ - Featurev0.9.0
Enable Fast Mode for Priority Models
Added a /fast toggle to route requests through priority queues for OpenAI and Anthropic models, significantly reducing latency for supported models like GPT-5.4 and Claude.
Source ↗ - Featurev0.9.0
Introduce Local Web Dashboard
Manage your Hermes Agent via a new browser-based interface. Configure settings, monitor active sessions, browse skills, and manage your gateway without needing to edit config files or use the terminal.
Source ↗ - Feature
Hermes Agent v0.8.0 Adds Notify-on-Complete, Live Model Switching
Hermes Agent v0.8.0 (April 8, 2026) adds notify_on_complete for background tasks, live /model switching mid-session, native Google AI Studio provider, MCP OAuth 2.1 PKCE, and OSV malware scanning.
Source ↗ - Feature
Hermes Agent v0.7.0 Adds Pluggable Memory, Credential Pools
Hermes Agent v0.7.0 introduces extensible memory plugins and auto-rotating credentials. Integrate Camofox browser for stealth tasks.
Source ↗ - Feature
Hermes Agent v0.6.0 Ships Profiles, MCP Mode, Docker Container
Hermes Agent v0.6.0 (March 30, 2026) ships profiles for multiple isolated instances, MCP server mode for Claude Desktop and Cursor, an official Docker image, and ordered fallback provider chains.
Source ↗ - Feature
Hermes Agent v0.5.0 Supports 400+ Models, Hardens Supply Chain
Hermes Agent v0.5.0 (March 28, 2026) routes 400+ models via Nous Portal, adds Hugging Face as a first-class provider, removes the compromised litellm dependency, and exposes plugin lifecycle hooks.
Source ↗ - Feature
Hermes Agent v0.4.0 Adds OpenAI API, 6 Messaging Platforms
Hermes Agent v0.4.0 (March 24, 2026) adds an OpenAI-compatible /v1/chat/completions endpoint, six new messaging adapters (Signal, DingTalk, SMS, Mattermost, Matrix, Webhook), @file/@url context, and gateway prompt caching.
Source ↗