Built-in Tools and the Nous Tool Gateway

9 min read Module 3 of 9 Topic 8 of 25

What you'll learn

  • List the major categories of built-in tools available in a default Hermes installation
  • Explain the difference between a Tool Gateway subscription tool and a bring-your-own-key tool
  • Configure optional tool API keys for premium capabilities like Firecrawl or ElevenLabs
  • Inspect and selectively enable built-in tools for a given deployment
Building this at your company? For enterprise and company teams taking this to production: book a 30-minute session with our AI engineers for architecture guidance, code review, and a rollout plan for your use case.
Book a Team Session

What Ships by Default

A standard Hermes installation includes 60+ built-in tools, grouped into toolsets you saw referenced in Lesson 7.3. The major categories:

CategoryExample toolsTypical use
Filesystemread/write files, search directory treesCodebase work, document editing
Shell / executionrun commands, git operationsDevelopment tasks, diagnostics
Websearch, fetch/scrape a URL, browser automationResearch, up-to-date information
Code executionsandboxed Python/JS runnerData analysis, calculations, quick scripts
Mediaimage generation, text-to-speech, transcriptionContent creation, voice interfaces (Module 6)
Schedulingcron-style recurring tasksReminders, periodic checks (Lesson 17)

Inspect what’s currently active in your installation:

hermes tools

The Nous Tool Gateway

Many of the more expensive or rate-limited capabilities, web search, browser automation, image generation, premium TTS, are unified behind the Nous Tool Gateway, accessible with a single Nous Portal subscription rather than five separate provider accounts. This is the tool-layer equivalent of the model-layer convenience Nous Portal offers in Lesson 4.

hermes gateway status   # confirm the gateway is reachable and which tools it's serving

Bring-Your-Own-Key Alternatives

If you’d rather not go through the Tool Gateway for a specific capability, or don’t have a Portal subscription, most tool categories have a direct provider option:

# ~/.hermes/.env

FIRECRAWL_API_KEY=...        # web scraping, direct
FIRECRAWL_API_URL=...        # optional: self-hosted Firecrawl instance

BROWSERBASE_API_KEY=...      # browser automation, direct
BROWSERBASE_PROJECT_ID=...

FAL_KEY=...                  # image generation, direct

ELEVENLABS_API_KEY=...       # premium TTS, direct
VOICE_TOOLS_OPENAI_KEY=...   # OpenAI TTS as an alternative

This mirrors the model-provider pattern exactly: Tool Gateway is the convenient default, direct keys are the escape hatch when you need a specific provider, want to avoid a subscription, or are running a deployment with strict vendor requirements.

Selectively Enabling Tools for a Deployment

Combine what you learned in subtopic 7.3 with the categories above to scope a deployment tightly. A research-only agent, for instance, has no reason to load code execution or media tools:

agent:
  platform_toolsets:
    cli: ["web_search", "filesystem"]
  disabled_toolsets: ["code_execution", "image_generation", "tts"]

A coding assistant flips that around: filesystem, shell, and code execution enabled, media tools disabled. There is no universally “correct” toolset, the right list is a direct function of what the deployment is actually for, which is also the central question Module 7’s guardrails material returns to.

Exercise: run hermes tools, pick three tools you don’t recognize, and read their descriptions (this is the same <tools> schema text from Lesson 7.1, just for tools Hermes ships rather than ones you wrote). For one of them, decide whether you’d want it enabled in a public-facing messaging bot (Module 6) or not, and why.

Knowledge Check

3 questions to test your understanding

1 What does the Nous Tool Gateway unify access to?

2 You want image generation but don't have a Nous Portal subscription. What's the alternative path?

3 A deployment only ever needs filesystem and shell tools, never web search or image generation. What did Lesson 7's subtopic 7.3 recommend doing, and why does it matter more with 60+ built-in tools available?

Go further with expert guidance

Ready to build production AI?
Talk to our R&D team.

These courses give you the foundation. Our embedded AI teams take you from prototype to production in 30–90 days, with your team, your codebase, your goals. Book a free strategy call to see how we can accelerate your AI initiative.

30 minutes · No obligation · Expert AI engineers, not sales reps

AI Architecture Review

Audit your current stack and identify high-impact improvements

Project Review

Get expert feedback on your AI implementation and codebase

Team Mentoring

Upskill your engineers with hands-on AI coaching sessions

AI Strategy

Define your AI roadmap, prioritization, and implementation plan