Latest Updates
News, releases, and stories from the Hermes Agent community.
Hermes Agent · May 19, 2026 · 8 min read From OpenClaw to Hermes Agent: a step-by-step migration
If you've been running OpenClaw, Hermes Agent is its successor. v0.14.0 ships a built-in `hermes claw migrate` that imports your SOUL.md, memories, skills, command allowlist, messaging configs, API keys, and TTS assets. This is the practical walkthrough — what gets imported, what flags you have, and the dry-run-first workflow that avoids surprises.
Hermes Agent · May 18, 2026 · 7 min read One Claude Pro subscription, every tool: using `hermes proxy` with Aider, Cline, and Codex CLI
v0.14.0 shipped a one-line command that quietly absorbed a whole class of integration questions: `hermes proxy`. Run it once and you get a local `http://localhost:port` endpoint that speaks the OpenAI API, backed by whichever OAuth provider you're signed into — Claude Pro, ChatGPT Pro, SuperGrok. Aider, Cline, Codex CLI, and Continue all expect an OpenAI-compatible endpoint. They just work with your existing subscription, no API key required. Here's the exact workflow, step by step.
Hermes Agent · May 18, 2026 · 9 min read From 7 to 22 in 14 months: every chat platform Hermes Agent runs on
Hermes Agent launched in March 2026 speaking seven chat platforms from a single gateway. Fourteen months later it speaks 22 — from Telegram and Discord to Microsoft Teams, LINE, and SimpleX Chat. Same memory, same skills, same agent across all of them. This is the complete map: every platform by category, the release each one landed in, and how the gateway architecture makes 'one agent, twenty-two front doors' possible.
Hermes Agent · May 17, 2026 · 9 min read Touring the 9 new skills that landed with Hermes Agent v0.14.0
v0.14.0 added nine new optional skills covering perp trading, market data, REST/GraphQL debugging, multi-chain EVM, evolutionary prompt tuning, OSINT, public-internet tunneling, change-detection polling, and a Notion overhaul. Plus huggingface/skills became a trusted default tap in the Skills Hub. This is the tour: what each skill does, when it's worth installing, the upstream PR for each.
Hermes Agent · May 17, 2026 · 10 min read How Hermes Agent v0.14.0 cut cold start by 19 seconds and made browser ops 180× faster
v0.14.0's performance wave is a stack of small wins that compound: cache the skills index, defer heavy imports, parallelize doctor checks, hold a persistent Chrome DevTools connection across browser calls. Net result: ~19 seconds off `hermes` cold start, 180× faster `browser_console`, and a cross-session prompt cache that survives `/new`. This is the PR-by-PR breakdown of how it actually got done — and the lessons that generalize.
Hermes Agent · May 17, 2026 · 11 min read Hermes Agent's security model: container isolation, command approval, and what's not protected
An autonomous agent that can run shell commands is one bad prompt injection away from destroying your filesystem. Hermes's security model is layered: container isolation as the primary boundary, a command-approval workflow for dangerous operations, OAuth scoping for messaging platforms, and (new in v0.14.0) tool error sanitization plus sudo brute-force blocking. Here's what each layer does, where the seams are, and what isn't protected.
Hermes Agent · May 17, 2026 · 8 min read Hermes Cron in practice: a daily report delivered to Telegram at 8 AM
Hermes Agent's cron scheduler turns a chat agent into a recurring agent — describe a job in natural language, set a schedule, and it runs unattended. With `deliver=all` the output fans out to every connected platform at once. This walkthrough builds a daily-report job: summarize overnight email, GitHub notifications, and calendar; deliver to Telegram at 8 AM weekdays. Same pattern works for nightly backups, weekly audits, on-call rotation pings.
Hermes Agent · May 17, 2026 · 9 min read Hermes Agent's 7 sandbox backends: when to pick which
When the agent runs `rm -rf` it should land in a container, not on your laptop. Hermes ships seven terminal-execution backends — local, Docker, SSH, Singularity, Modal, Daytona, and Vercel Sandbox — each with different cost, latency, and isolation tradeoffs. This is the decision guide: when each one is the right answer, what they cost, and how to switch between them.
Hermes Agent · May 17, 2026 · 8 min read Let Hermes run until it's done: `/goal` and `/subgoal` in practice
Most AI agents run one turn at a time and stop when you stop typing. Hermes Agent's `/goal` flips that: you set a target plus success criteria, the agent loops — propose, execute, judge — until a separate judge LLM agrees the criteria are met. v0.14.0 added `/subgoal` so you can layer extra criteria onto an active loop mid-flight. Here's what's actually happening under the hood, when this beats turn-by-turn, and when it bites you.
Hermes Agent · May 16, 2026 · 11 min read Hermes Agent v0.14.0 — 22 highlights you shouldn't miss
May 16, 2026: Hermes Agent shipped v0.14.0 — 633 merged PRs, 165,061 insertions, 215 contributors in a nine-day window. This is the comprehensive recap, 22 highlights organized by what they do for you: from `pip install hermes-agent` (finally) and `hermes proxy` (turning your Claude Pro into a local OpenAI endpoint), to Microsoft Teams end-to-end, 19 seconds shaved off cold start, 180× faster browser operations, and nine new optional skills. Every claim is cited back to the upstream release notes.