Configuration File
Posit Assistant is configured through a JSON settings file. This file controls model selection, behavior preferences, runtime configuration, permissions, and more.
File Locations
Posit Assistant uses two levels of configuration:
-
Global config applies to all projects. Located at:
- macOS / Linux:
~/.positai/settings.json - Windows:
%USERPROFILE%\.positai\settings.json
- macOS / Linux:
-
Project config overrides global settings for a specific project. Place a
.positai/settings.jsonfile in your project root.
Some settings (like providers, storage, and logging) are global-only. Others (like model, permissions, and MCP servers) can be set at either level, with project values taking precedence. The sections below indicate which level each setting supports.
Settings are resolved in this order (last wins): defaults → global config → project config → environment variables.
Per-Project Settings
The following settings can be set in either the global ~/.positai/settings.json or a project-level .positai/settings.json. Project values take precedence over global values.
Model Settings
| Key | Type | Default | Description |
|---|---|---|---|
model.id | string | "claude-sonnet-4-6" | The model to use for conversations. |
model.provider | string | "positai" | The LLM provider. The default is "positai" (Posit AI — a managed service from Posit). See the Posit AI page for details. |
model.thinkingEffort | string | — | Default thinking effort level for new conversations. During a conversation, thinking effort is adjusted via the UI and stored per-conversation. |
model.webSearch | boolean | false | Default web search preference for new conversations. During a conversation, web search is toggled via the UI and stored per-conversation. |
Permissions
Control which tools the assistant can use. Each key is a tool name (bash, read, edit, etc.) mapped to either an action string or an object of glob patterns:
{
"permission": {
"bash": {
"git *": "allow",
"rm *": "deny"
},
"read": "allow",
"edit": "ask"
}
}
You can also set a single action for all tools as a shorthand:
{
"permission": "ask"
}
| Key | Type | Default | Description |
|---|---|---|---|
permission | string | object | — | Tool permission configuration. Set to "allow", "ask", or "deny" for all tools, or an object with per-tool rules. Pattern matching uses glob syntax for shell commands. |
When you grant or deny permissions through the UI, they are saved to your project-level
.positai/settings.json. If you want permissions to apply across all projects, copy the
permission block to your global ~/.positai/settings.json instead.
MCP Servers
Connect external tool servers using the Model Context Protocol.
{
"mcpServers": {
"my-server": {
"type": "local",
"command": ["node", "server.js"],
"environment": {
"API_KEY": "..."
}
},
"remote-server": {
"type": "remote",
"url": "https://mcp.example.com",
"headers": {
"Authorization": "Bearer ..."
}
}
}
}
| Key | Type | Default | Description |
|---|---|---|---|
type | string | "local" | Server type. "local" for stdio subprocess, "remote" for HTTP/SSE. |
command | string[] | — | Command and arguments to start a local MCP server. |
url | string | — | URL of a remote MCP server. |
enabled | boolean | true | Whether this server is active. |
timeout | number | 10000 | Connection timeout in milliseconds. |
MCP servers can be configured in either the project-level .positai/settings.json or the global
~/.positai/settings.json. Project-level servers are merged with global servers by server name,
with project values taking precedence. Use project-level config for project-specific servers and
global config for servers you want available everywhere.
Skills
Control where Posit Assistant discovers custom skills.
| Key | Type | Default | Description |
|---|---|---|---|
skills.paths | string[] | ["~/.agents/skills", "~/.positai/skills", ".agents/skills", ".positai/skills"] | Directories to search for skills, processed in order. Paths starting with ~ or $HOME are expanded to the home directory. Relative paths are resolved against each workspace root. |
When the same skill name exists in more than one directory, the version from the later directory is used.
When both global and project-level configs define skills.paths, the project value
completely replaces the global value — paths are not merged. To extend the defaults, repeat them
in your project config alongside any additions.
Runtime Settings
Control how Posit Assistant connects to R and Python.
| Key | Type | Default | Description |
|---|---|---|---|
runtime.r.enabled | boolean | true | Enable R runtime integration. |
runtime.r.path | string | "Rscript" | Path to the Rscript executable. |
runtime.r.timeout | number | 30000 | Timeout in milliseconds for R code execution. |
runtime.python.enabled | boolean | true | Enable Python runtime integration. |
runtime.python.path | string | "python3" | Path to the Python executable. |
runtime.python.timeout | number | 30000 | Timeout in milliseconds for Python code execution. |
Workspace Settings
| Key | Type | Default | Description |
|---|---|---|---|
workspace.path | string | "" | Working directory. Defaults to the current directory. |
workspace.allowedRoots | string[] | — | Allowed root directories for workspace switching. Defaults to the user home directory. |
Sandbox
Control sandbox mode for shell tool execution. On macOS and Linux, bash commands run inside an OS-level sandbox (macOS Seatbelt / Linux bubblewrap) that restricts writes to the workspace and temp directories, blocks network access, and prevents reading sensitive paths like ~/.ssh and ~/.aws. On Windows, the same setting enables gated mode for the active shell: commands are checked against a built-in allowlist before they run. This is not an OS-level sandbox.
| Key | Type | Default | Description |
|---|---|---|---|
sandbox.enabled | boolean | false | Enable sandboxing for shell tool execution. |
You can also toggle sandbox mode during a session with the /sandbox command.
Compaction
Control how Posit Assistant manages long conversations. See Context Management for details on how compaction works.
| Key | Type | Default | Description |
|---|---|---|---|
autoCompactTokenBuffer | number | 30000 | Tokens reserved for the compaction summary. Auto-compaction triggers once fewer than this many tokens remain in the context window. |
microCompactTokenThreshold | number | 150000 | Token count above which micro-compaction activates, silently trimming old tool results. |
microCompactKeepRecentCount | number | 8 | Number of recent user turns preserved without micro-compaction. |
Global Settings
The following settings can only be set in the global ~/.positai/settings.json. They are not supported in project-level config files.
Provider Settings
Provider credentials and connection details are configured under the providers key. Credentials are normally set through the application’s settings UI, but you can override the base URL or attach custom HTTP headers in the config file. This is useful for enterprise proxies or custom API gateways that require extra tenancy or routing markers (e.g. Databricks’ x-databricks-use-coding-agent-mode).
{
"providers": {
"positai": {
"baseUrl": "https://my-proxy.example.com",
"customHeaders": {
"x-gateway-tenant": "team-42"
}
}
}
}
| Key | Type | Default | Description |
|---|---|---|---|
providers.{provider}.baseUrl | string | — | Custom base URL for the provider's API. |
providers.{provider}.customHeaders | object | — | Extra HTTP headers attached to every request for this provider. Useful for enterprise gateways that require extra tenancy or routing markers. Do not put SDK-managed header names (Authorization, x-api-key, anthropic-version) here — on chat requests they would clobber the SDK's auth headers and break the connection. |
providers.allowedProviders | string[] | — | Restrict which providers are available. If omitted, the platform default is used. |
In Positron, base URLs and custom headers are stored in Positron’s authentication.* settings.
See Positron Settings for details.
Google Vertex AI
For Google Vertex AI on non-Positron platforms, configure your Google Cloud project and Vertex AI location, then authenticate with Application Default Credentials:
{
"providers": {
"google-vertex": {
"googleCloud": {
"project": "my-gcp-project",
"location": "us-central1"
}
}
}
}
In Positron, configure Google Vertex AI through the provider configuration dialog instead of editing ~/.positai/settings.json directly.
Storage Settings
| Key | Type | Default | Description |
|---|---|---|---|
storage.path | string | "~/.positai" | Directory for conversation storage. |
Logging
| Key | Type | Default | Description |
|---|---|---|---|
logging.level | string | "info" | Log level. Options: "error", "warn", "info", "debug", "trace". |
logging.file | string | "" | Log file path. Defaults to ~/.positai/logs/{platform}.log. |
logging.console | boolean | true | Output logs to the console. |
Example Configuration
{
"model": {
"provider": "positai",
"id": "claude-sonnet-4-6"
},
"permission": {
"read": "allow",
"bash": {
"git *": "allow"
}
},
"skills": {
"paths": [
"~/.agents/skills",
"~/.positai/skills",
".agents/skills",
".positai/skills",
"~/my-company/shared-skills"
]
}
}
Some configuration changes require restarting the assistant or opening a new conversation to take effect.
Environment Variables
Some settings can also be controlled via environment variables. Environment variables take precedence over file-based configuration.
| Variable | Description |
|---|
| GOOGLE_CLOUD_PROJECT | Google Cloud project for the Vertex AI provider |
| GOOGLE_CLOUD_LOCATION | Google Cloud location for the Vertex AI provider |
| POSITAI_BASE_URL | Custom base URL for the Posit AI provider |
Never commit API keys to version control. Use environment variables or a secrets manager for sensitive credentials.