This page explains where Oppla stores its configuration, how to edit settings (GUI and file-based), and provides practical examples for common tasks referenced elsewhere in the docs (key bindings, themes, AI provider setup). If you’re coming from the Getting Started or Key Bindings guide, this is the canonical reference for configuring Oppla.

​
Where settings live

Oppla stores configuration at user and project scope. Use the GUI preferences for most tasks or the settings files for automation and reproducibility.
  • User settings (per-user)
    • macOS / Linux: ~/.config/oppla/settings.json
    • Windows (planned): %APPDATA%\Oppla\settings.json
  • Project settings (per-repository)
    • Repository root: .oppla/settings.json
    • Project settings override user settings for that repository.
Oppla also uses OS-native secret storage for API keys and sensitive tokens (Keychain, Secret Service, etc.). See the Privacy & Security page for details.

​
Open settings

  • GUI: Preferences → Settings (or Command Palette → oppla: Open Settings)
  • File: Open ~/.config/oppla/settings.json (or project .oppla/settings.json) in the editor
Note: Settings can be provisioned by dotfiles or CI when onboarding developers.

​
Common configuration areas

​
AI providers and models

Configure cloud or local AI providers, preferred models, and privacy options. Use environment variables for secrets (recommended). Example (user settings snippet): 
{
  "ai": {
    "provider": "openai",
    "providers": {
      "openai": {
        "api_key_env": "OPENAI_API_KEY",
        "default_model": "gpt-4o-mini",
        "timeout_ms": 30000
      },
      "ollama": {
        "endpoint": "http://localhost:11434",
        "default_model": "llama2-13b-q4"
      }
    },
    "privacy": {
      "mode": "local_first",
      "send_code": "opt_in"
    }
  }
}
Key notes:
  • Use api_key_env to point to env vars rather than storing keys in files.
  • privacy.mode can be local_only, local_first, or cloud_allowed.
  • Configure task-specific models in task_model_overrides (see Available Models page).

​
Key bindings / keymaps

Oppla reads user keymaps from ~/.config/oppla/keymap.json. You can set a base_keymap in settings to adopt a familiar layout (VSCode, Emacs, Vim, etc.). Example: 
{
  "keymap": {
    "base_keymap": "VSCode",
    "vim_mode": false,
    "ai_keymap_learning": {
      "enabled": true,
      "suggest_after_days": 7
    }
  }
}
  • Open default keymap: Command Palette → oppla: open default keymap
  • For debugging: run dev: Open Key Context View
  • To enable AI-optimized learning, allow ai_keymap_learning.enabled.
See the Key Bindings guide for syntax and examples.

​
Themes & visual customization

Theme preferences live in settings; Oppla supports ai_adaptive mode that auto-adjusts based on time of day and activity. Example: 
{
  "theme": {
    "mode": "ai_adaptive",
    "light": "Oppla Light",
    "dark": "Oppla Dark",
    "experimental.theme_overrides": {
      "editor.background": "#0f1724",
      "ai.suggestion.background": "#10203a"
    }
  }
}
To load local themes, place theme JSON files under ~/.config/oppla/themes/.

​
CLI settings & installation

  • CLI binary (when installed) integrates with your system shell.
  • To install CLI on macOS: oppla: cli install (creates a symlink to /usr/local/bin/oppla)
  • For automated installs, prefer signed release assets and verify checksums.
Security note: Avoid piping remote install scripts directly (curl | sh) without verifying signatures in production environments.

​
Secret management & environment variables

Preferred approaches:
  • Local dev: store secrets in OS secret stores (Oppla prompts to save keys securely).
  • CI: set provider keys as environment variables (OPENAI_API_KEY, ANTHROPIC_API_KEY, etc.).
  • Project automation: use a secrets manager and reference tokens with api_key_env.
Example environment variable reference in settings: 
"providers": {
  "openai": { "api_key_env": "OPENAI_API_KEY" }
}

​
Project-level configuration and AI Rules

Project-level AI behavior and enforcement (AI Rules) should live in .oppla/ai-rules.json in the repository root. Use project rules to:
  • Prevent sending sensitive files to cloud providers
  • Require approvals for high-risk paths
  • Restrict allowed providers or models
Rules are applied before agent runs or outbound model requests.

​
Troubleshooting

  • Settings not applied:
    • Confirm you’re editing the correct scope (user vs. project).
    • Restart Oppla or reload the window after manual edits.
  • API key errors:
    • Verify env var is set and contains a valid key.
    • Check OS secret store permissions and that Oppla can read the secret.
  • Keymap conflicts:
    • Use dev: Open Key Context View and run “AI: Analyze Keymap Usage”.
  • Slow AI responses:
    • Switch to local_first, reduce context window size, or pick a lower-latency local model.

​
Best practices checklist

  • Use OS secret stores or environment variables for API keys.
  • Start AI keymap learning with a default period (7 days) before applying suggestions.
  • Prefer local-first or local-only mode for sensitive codebases.
  • Keep project-level rules under source control for reproducibility.
  • Document any non-standard settings in the repository README for contributors.
  • Key Bindings: docs/ide/configuration/key-bindings.mdx
  • Themes: docs/ide/configuration/themes.mdx
  • AI Overview & configuration: docs/ide/ai/overview.mdx, docs/ide/ai/configuration.mdx
  • Privacy & Security: docs/ide/ai/privacy-and-security.mdx
  • Agent Panel, Rules, Tools: docs/ide/ai/agent-panel.mdx, docs/ide/ai/rules.mdx, docs/ide/ai/tools.mdx
If you want, I can:
  • Add step-by-step screenshots for common configuration flows.
  • Provide example automation scripts to provision settings for teams.
  • Add validation rules and a sample oppla: validate-config CLI snippet to check config health.