Compatible (BYOK)Use Hermes Agent with GlideflowAI.
A self-directed personal agent with custom endpoint support. Connect it through GlideflowAI's OpenAI-compatible API with a key you control.
What this connection is for
Hermes Agent is a personal agent project from Nous Research built for persistent, tool-using workflows. Rather than functioning only as a chat box, it can be configured around local skills, tools, and recurring work patterns. That makes it compelling for technically confident users who want an agent that can gradually become familiar with a personal workflow—but it also makes safe secrets handling and permission boundaries especially important.
The project documents custom OpenAI-compatible endpoints through an API key and base URL. In a personal-agent setup, however, a successful connection is only the beginning: the agent may have access to files, shells, browsers, or integrations. Start with the smallest set of capabilities, a distinct low-privilege key where possible, and an environment you can inspect. Never confuse a compatible model endpoint with a blanket safety guarantee for the rest of an agent system.
Hermes Agent is listed Compatible (BYOK) because its documented custom endpoint variables align with GlideflowAI's protocol, while this gallery does not claim an end-to-end validation of every Hermes release and skill stack. The connection belongs to the user; it does not indicate an endorsement, collaboration, or support commitment from Nous Research.
Why pair it with GlideflowAI
A custom endpoint gives a personal-agent environment access to a single gateway key alongside your other compatible tools.
GlideflowAI's public USD pricing and catalog can help you choose a model deliberately before a multi-step agent workflow begins.
Setup
Use a new API key for the tool, select a model from the catalog, and begin with a small task. Do not put a key in a repository, screenshot, or shared configuration file.
Use a private environment configuration
Follow the current Hermes Agent configuration reference and set the custom OpenAI-compatible variables in its supported environment or secret store. Do not paste keys into skills, prompts, or repositories.
Set the endpoint and API key
Hermes documents OPENAI_BASE_URL and OPENAI_API_KEY for a custom endpoint. Use the gateway's /v1 URL and your key.
export OPENAI_BASE_URL="https://api.glideflowai.com/v1" export OPENAI_API_KEY="sk-your-key"Select a model and restrict the first task
Choose an exact supported model in Hermes's current provider configuration, then test a low-risk task with minimal tool permissions before connecting personal services.
Before you begin: Manual check: confirm the active Hermes Agent release's model-selection setting in addition to its documented OPENAI_BASE_URL and OPENAI_API_KEY variables.
Need the endpoint basics? Read authentication, browse the model catalog, or compare pricing.
Frequently asked questions
Why is Hermes Agent not Verified?
Its custom endpoint mechanism is documented, but GlideflowAI has not claimed a complete test across Hermes versions, skills, and personal-agent deployments.
Can an agent key be stored in a skill?
No. Store it in Hermes's supported secret or environment configuration, never in a prompt, skill, or checked-in file.
What should I secure first?
Begin with least-privilege tool access and an isolated test task; a valid API connection should not expand the agent's permissions.
Ready to connect Hermes Agent?
Create a key, use the setup above, and verify the connection on a small task before you scale it up.
