How do I install Hal (http Api Layer)?
The recommended method is `npx hal-mcp`. Alternatively, install globally with `npm install -g hal-mcp` and then run `hal-mcp`.
@DeanWard
Basic information
HAL is a Model Context Protocol (MCP) server that provides HTTP API capabilities to Large Language Models. It allows LLMs to make HTTP requests and interact with web APIs through a secure, controlled interface. HAL can also automatically generate tools from OpenAPI/Swagger specif
Add this server to your MCP-compatible client using the configuration below.
{
"mcpServers": {
"hal": {
"command": "npx",
"args": [
"hal-mcp"
]
}
}
}Make an HTTP GET request to a specified URL. Supports secret substitution using {secrets.key} syntax where 'key' corresponds to HAL_SECRET_KEY environment variables.
Make an HTTP POST request to a specified URL with optional body and headers. Supports secret substitution using {secrets.key} syntax in URL, headers, and body where 'key' corresponds to HAL_SECRET_KEY environment variables.
Make an HTTP PUT request to a specified URL with optional body and headers. Supports secret substitution using {secrets.key} syntax in URL, headers, and body where 'key' corresponds to HAL_SECRET_KEY environment variables.
Make an HTTP PATCH request to a specified URL with optional body and headers. Supports secret substitution using {secrets.key} syntax in URL, headers, and body where 'key' corresponds to HAL_SECRET_KEY environment variables.
Make an HTTP DELETE request to a specified URL with optional headers. Supports secret substitution using {secrets.key} syntax in URL and headers where 'key' corresponds to HAL_SECRET_KEY environment variables.
Make an HTTP HEAD request to a specified URL with optional headers (returns only headers, no body). Supports secret substitution using {secrets.key} syntax in URL and headers where 'key' corresponds to HAL_SECRET_KEY environment variables.
Make an HTTP OPTIONS request to a specified URL to check available methods and headers. Supports secret substitution using {secrets.key} syntax in URL and headers where 'key' corresponds to HAL_SECRET_KEY environment variables.
Get a list of available secret keys that can be used with {secrets.key} syntax. Only shows the key names, never the actual secret values.
Hal (http Api Layer) is a Model Context Protocol (MCP) server that gives Large Language Models the ability to make HTTP requests and interact with web APIs through a secure, controlled interface. It supports common HTTP methods, includes built-in secret management using environment variables, and can automatically generate tools from OpenAPI/Swagger specifications.
Install Hal globally via npm install -g hal-mcp or run it on demand with npx hal-mcp. Configure it in MCP-compatible clients (e.g., Claude Desktop) by adding an entry to the mcpServers configuration object. Optionally set environment variables like HAL_SWAGGER_FILE and HAL_SECRET_* to enable Swagger integration and secure token management.
{secrets.key} template substitution.list-secrets, http-get, and http-post.The recommended method is `npx hal-mcp`. Alternatively, install globally with `npm install -g hal-mcp` and then run `hal-mcp`.
Define secrets as environment variables with the `HAL_SECRET_` prefix (e.g., `HAL_SECRET_API_KEY=abc`). In HTTP requests, use `{secrets.api_key}` as a placeholder. The AI never sees the actual secret; it is substituted at request time.
Yes. Set the `HAL_SWAGGER_FILE` environment variable to the path of your OpenAPI/Swagger file (JSON or YAML). Optionally override the base URL with `HAL_API_BASE_URL`. Hal will then create tools for each endpoint.
Use namespaces and `HAL_ALLOW_*` environment variables. For example, `HAL_SECRET_MICROSOFT_API_KEY=key` and `HAL_ALLOW_MICROSOFT="https://azure.microsoft.com/*"` ensures that secret is only substituted in requests matching those patterns.
Hal provides tools for GET, POST, PUT, PATCH, DELETE, OPTIONS, and HEAD requests through built-in tools like `http-get` and `http-post`.
Set up and manage the Eddyter AI rich text editor by talking to your AI agent — for Claude, Cursor, Codex and other MCP tools. A Model Context Protocol server that lets AI tools add Eddyter to a React, Next.js, Vite or R
AptiBuild AI — Career Intelligence MCP Server Structured career decision scaffolding for AI agents. Provides personality-conditioned business idea generation, official BLS OEWS wage data, FRED labor market signals, and
CronWatch records every run of your scheduled jobs in your own database, and speaks up when one is missed, fails, gets stuck, runs slow or costs too much. CronWatch is free and open-source.

Agents make claims. Reelier writes receipts — record an agent's tool-call workflow once, replay it deterministically at 0 tokens, and diff runs to catch drift.

Local-first credential broker for MCP. AI coding agents (Claude Code, Cursor, Codex…) call APIs, SSH and databases through scoped presets; secrets stay in your OS keychain and never enter the agent's context.
Comments