skills.rest
lurodrisilva
lurodrisilvaCOMMUNITY
Explore all skills by this author →
GitHub

hermes-agent

Install, configure, operate, and extend the Hermes Agent autonomous AI framework.

Updated Apr 11, 2026
Search Tags:#MCP#AI framework#autonomous agent#llm providers#hermes-agent#cron automation#messaging gateway
One-click install
npx skills add https://github.com/lurodrisilva/personal-skills --skill hermes-agent-lurodrisilva
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: hermes-agent
Source: https://github.com/lurodrisilva/personal-skills/tree/main/ai/hermes-agent
Command: npx skills add https://github.com/lurodrisilva/personal-skills --skill hermes-agent-lurodrisilva

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Operating the Hermes Agent framework involves a dense set of non-obvious contracts — a script-only install, a strict secrets-vs-config split, a 64k-token model context floor, profile isolation rules, and safety rails that shift between terminal backends — and getting any of them wrong breaks the agent or leaks credentials. ## Core Features & Use Cases - Full lifecycle playbook: Six phased guides covering install/operate, config/profiles/secrets/egress, skills/memory/context, automation/gateway/serving, providers/MCP/Python library, and extending the framework with plugins and tools. - Read-only triage scripts: Three bash scripts audit the install and config split, inventory profiles/skills/cron/MCP/plugins/memory, and probe the dashboard (9119), API server (8642), gateway, and iron-proxy egress status without mutating anything. - Use Case: You want to run a scheduled nightly backlog triage agent on Telegram. Use this Skill to create an isolated profile, configure cron with delivery targets, wire the messaging gateway with default-deny authorization, and verify everything with the surface probe script. ## Quick Start Ask the AI to audit your local Hermes Agent installation and config split using the hermes-env-audit script, then walk you through fixing any warnings.

Frequently Asked Questions about hermes-agent

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I install Hermes Agent on Linux or macOS?▼

Install Hermes Agent only via the official script: curl -fsSL https://hermes-agent.nousresearch.com/install.sh | bash, or install.ps1 on Windows. pip, uv tool, Homebrew, AUR, and macOS Intel are explicitly unsupported; Git is the only universal prerequisite.

How do I schedule automated agent tasks with Hermes cron?▼

Use hermes cron create with a schedule like "every 2h" or a 5-field cron expression, a self-contained prompt, and a delivery target such as telegram or slack. Script jobs with --no-agent poll for $0, and a final line of {"wakeAgent": false} skips the LLM entirely.

Where do API keys go in Hermes Agent config?▼

Secrets go only in ~/.hermes/.env with chmod 600 permissions; behavior goes only in config.yaml via hermes config set. Never hand-edit the YAML or put key-shaped values in it — profile exports and distributions always strip .env and auth.json.

Can Hermes Agent use local LLMs like Ollama or llama.cpp?▼

Yes, Hermes supports Ollama, llama.cpp, vLLM, SGLang, and LM Studio, but every model needs at least 64,000 tokens of context. Tool-calling must be wired explicitly, for example llama.cpp with --jinja -c 64000 or vLLM with --enable-auto-tool-choice --tool-call-parser hermes.

Why is it unsafe to run two Hermes agents on one profile?▼

A profile is a separate Hermes home with isolated config, memories, sessions, and state.db, so two agent processes writing to the same profile compound each other's memory into unauthored state. Create separate profiles with hermes profile create instead.

What are the limitations of the Hermes iron-proxy egress firewall?▼

Iron-proxy works only with the Docker terminal backend in v1, and AWS SigV4 (Bedrock) plus GCP OAuth (Vertex) bypass it entirely. After token rotation, running sandboxes keep old tokens until restarted, and it never binds 0.0.0.0.

Related Skills

Explore More Software Engineering Skills→
thinking-machines-labthinking-machines-lab
4.0k

setup

Guide Tinker installation, environment setup, API key provisioning, and verification.

Official
Intermediate
i-bebsii-bebsi

hermes-agent

Configure, operate, and extend the Hermes Agent CLI and multi-platform gateway.

Community
Advanced
gke-labsgke-labs
42

review-security-k8s-agents-firewall

Audit Kubernetes network policies and firewall rules for AI agent sandboxes.

Official
Intermediate
dperussinadperussina

harness-selection

Decides whether to adopt an agent framework or build the harness for agent runtimes.

Community
Intermediate