hermes-provider-setup
starsforks
How to install hermes-provider-setup?
timfewi/tentaflake/hermes-provider-setup
$npx -y skills add timfewi/tentaflake --skill hermes-provider-setupConfigure AI providers for Hermes — Nous Portal, OpenRouter, Anthropic, OpenAI, custom endpoints, multi-provider fallback, credential pools
SKILL.md
| 1 | # Hermes Provider Setup |
| 2 | |
| 3 | > **Tentaflake context:** Inside a tentaflake agent container, you can run |
| 4 | > `hermes config set` interactively — but if your agent was defined with a |
| 5 | > `settings` attrset in `my-agents.nix`, the `config.yaml` inside the |
| 6 | > container is **read-only** (mounted `:ro`). Interactive changes are lost |
| 7 | > on restart. For a persistent configuration, put settings in `my-agents.nix` |
| 8 | > instead and rebuild. API keys always go in `/run/secrets/hermes-<name>.env` |
| 9 | > (or agenix), not via `hermes config set`. |
| 10 | |
| 11 | ## When to Use |
| 12 | |
| 13 | - Configure LLM provider for Hermes (first-time setup) |
| 14 | - Switch between providers |
| 15 | - Add multi-provider fallback |
| 16 | - Set up custom/self-hosted endpoint (Ollama, vLLM, SGLang) |
| 17 | - Configure credential pool for multi-key rotation |
| 18 | - Troubleshoot provider connectivity |
| 19 | - Switch models mid-session |
| 20 | |
| 21 | ## Procedure |
| 22 | |
| 23 | ### 1. Provider Selection Guide |
| 24 | |
| 25 | | Provider | Auth | Best for | |
| 26 | | -------------------- | ------------------- | ----------------------------------------------------------------- | |
| 27 | | **Nous Portal** | OAuth (browser) | All-in-one: 300+ models + Tool Gateway (web, image, TTS, browser) | |
| 28 | | **OpenAI Codex** | OAuth (device code) | Codex models, ChatGPT subscribers | |
| 29 | | **Anthropic** | OAuth or API key | Claude models (Max plan + credits) | |
| 30 | | **OpenRouter** | API key | Multi-provider routing, 200+ models | |
| 31 | | **Google AI Studio** | API key | Gemini models | |
| 32 | | **Custom Endpoint** | API key + base URL | Self-hosted (Ollama, vLLM, SGLang, LM Studio) | |
| 33 | | **DeepSeek** | API key | DeepSeek models direct | |
| 34 | | **AWS Bedrock** | IAM role | Claude, Nova within AWS | |
| 35 | | **GitHub Copilot** | OAuth | Copilot subscription models | |
| 36 | | **xAI** | API key or OAuth | Grok models | |
| 37 | | **z.ai / ZhipuAI** | API key | GLM models | |
| 38 | | **Kimi / Moonshot** | API key | Kimi models | |
| 39 | | **MiniMax** | API key or OAuth | MiniMax models | |
| 40 | |
| 41 | **Minimum context: 64K tokens.** Models with smaller windows rejected at startup. |
| 42 | |
| 43 | **Rule of thumb:** If Hermes can't complete a normal chat, don't add features yet. Get one clean conversation working. |
| 44 | |
| 45 | ### 2. Fastest Path — Nous Portal |
| 46 | |
| 47 | ```bash |
| 48 | hermes setup --portal |
| 49 | ``` |
| 50 | |
| 51 | This one command: |
| 52 | |
| 53 | 1. Opens browser for OAuth login (portal.nousresearch.com) |
| 54 | 2. Stores refresh token at `~/.hermes/auth.json` |
| 55 | 3. Lets you pick a model |
| 56 | 4. Sets Nous as inference provider |
| 57 | 5. Turns on Tool Gateway (web, image, TTS, browser) |
| 58 | 6. Ready to `hermes chat` |
| 59 | |
| 60 | After setup, config looks like: |
| 61 | |
| 62 | ```yaml |
| 63 | model: |
| 64 | provider: nous |
| 65 | default: anthropic/claude-sonnet-4.6 |
| 66 | base_url: https://inference-api.nousresearch.com/v1 |
| 67 | ``` |
| 68 | |
| 69 | ### 3. API Key Providers |
| 70 | |
| 71 | ```bash |
| 72 | # OpenRouter — most flexible multi-provider |
| 73 | hermes config set OPENROUTER_API_KEY sk-or-... |
| 74 | |
| 75 | # Anthropic — direct Claude access |
| 76 | hermes config set ANTHROPIC_API_KEY sk-ant-... |
| 77 | |
| 78 | # OpenAI |
| 79 | hermes config set OPENAI_API_KEY sk-... |
| 80 | |
| 81 | # Google Gemini |
| 82 | hermes config set GOOGLE_API_KEY AIza... |
| 83 | # or GEMINI_API_KEY |
| 84 | |
| 85 | # DeepSeek |
| 86 | hermes config set DEEPSEEK_API_KEY sk-... |
| 87 | |
| 88 | # xAI (Grok) |
| 89 | hermes config set XAI_API_KEY ... |
| 90 | |
| 91 | # AWS Bedrock — uses IAM role or aws configure |
| 92 | # No API key needed |
| 93 | |
| 94 | # GitHub Copilot |
| 95 | hermes config set COPILOT_GITHUB_TOKEN ghp_... |
| 96 | ``` |
| 97 | |
| 98 | ### 4. OAuth Providers |
| 99 | |
| 100 | ```bash |
| 101 | # Nous Portal |
| 102 | hermes portal # One-shot OAuth setup |
| 103 | hermes portal info # Check login status |
| 104 | hermes portal tools # Tool Gateway catalog |
| 105 | |
| 106 | # Anthropic OAuth |
| 107 | hermes auth add anthropic --type oauth |
| 108 | |
| 109 | # OpenAI Codex |
| 110 | hermes model → pick OpenAI Codex (device code flow) |
| 111 | |
| 112 | # Google Gemini OAuth |
| 113 | hermes model → pick Google Gemini (OAuth) |
| 114 | |
| 115 | # GitHub Copilot OAuth |
| 116 | hermes model → pick GitHub Copilot (OAuth) |
| 117 | |
| 118 | # MiniMax OAuth |
| 119 | hermes model → pick MiniMax (OAuth) |
| 120 | |
| 121 | # xAI Grok OAuth |
| 122 | hermes model → pick xAI Grok OAuth |
| 123 | |
| 124 | # Qwen OAuth |
| 125 | hermes model → pick Qwen OAuth |
| 126 | ``` |
| 127 | |
| 128 | ### 5. Custom Endpoint Configuration |
| 129 | |
| 130 | For local/self-hosted models (Ollama, vLLM, SGLang, LM Studio |
People who liked this also viewed
6 picksType
- microsoft-foundryDeploy, evaluate, fine-tune, and manage Foundry agents end-to-end with azd: hosted agent scaffold/run/deploy, prompt agent create, batch eval, continuous eval, prompt optimizer, Agent Optimizer scaffoSkillsJul 2026
- azure-deployExecute Azure deployments for ALREADY-PREPARED applications that have existing .azure/deployment-plan.md and infrastructure files. DO NOT use this skill when the user asks to CREATE a new application SkillsJul 2026
- azure-preparePrepare azd-based Azure projects for deployment: generates azure.yaml, infrastructure (Bicep/Terraform), and Dockerfiles for the Azure Developer CLI (azd) workflow. USE ONLY when the user explicitly wSkillsJul 2026
- azure-validate--- name: azure-validate description: "Pre-deployment validation for Azure readiness. Run deep checks on configuration, infrastructure (Bicep or Terraform), RBAC role assignments, managed identity pSkillsJul 2026
- azure-aigatewayConfigure Azure API Management as an AI Gateway for AI models, MCP tools, and agents. WHEN: semantic caching, token limit, content safety, load balancing, AI model governance, MCP rate limiting, jailbSkillsJul 2026
- azure-hosted-copilot-sdkOfficial agent plugin providing skills and MCP server configurations for Azure scenarios.SkillsJul 2026
