|
| 1 | +--- |
| 2 | +# GENERATED FILE — do not edit. Translations live in |
| 3 | +# src/ha_mcp/settings_ui/locales/cs.json (keys addon.*, features.*, |
| 4 | +# addon_dev.*); regenerate with: python scripts/generate_locales.py |
| 5 | +configuration: |
| 6 | + backup_hint: |
| 7 | + name: Full-HA snapshot suggestion level (LLM-facing) |
| 8 | + description: 'Tunes the wording of the hint the LLM sees in the |
| 9 | +
|
| 10 | + `ha_manage_backup(scope=''snapshot'')` tool description for when to |
| 11 | +
|
| 12 | + suggest a full Home Assistant tarball backup before risky operations |
| 13 | +
|
| 14 | + (e.g. mass device deletes). Affects ONLY this one LLM-facing prompt |
| 15 | +
|
| 16 | + sentence. |
| 17 | +
|
| 18 | +
|
| 19 | + For per-edit automatic snapshots of automations / scripts / helpers / |
| 20 | +
|
| 21 | + etc., use the separate **Enable auto-backup of edits** toggle below.' |
| 22 | + secret_path: |
| 23 | + name: Secret path override |
| 24 | + description: 'Optional custom HTTP path for the MCP server. Leave empty to use |
| 25 | + the auto-generated secure path. |
| 26 | +
|
| 27 | + ' |
| 28 | + enable_tool_search: |
| 29 | + name: Enable tool search |
| 30 | + description: 'Replace the full tool catalog with search-based discovery. Reduces |
| 31 | + idle context by roughly 90%, to about 5K tokens. ⚠️ Do NOT enable this in clients |
| 32 | + with their own built-in tool search / deferred tools (claude.ai, Claude Desktop, |
| 33 | + Claude Code) — the two search layers conflict, and the client''s built-in tool |
| 34 | + search is the better choice there; leave this off. Whether tools are deferred |
| 35 | + depends on your client and model combination: use this when your setup loads |
| 36 | + the full catalog up front — models without native deferred tools (e.g. Gemini, |
| 37 | + OpenAI-compatible local models, Claude Haiku), clients that inline all tool |
| 38 | + schemas regardless of model (e.g. GitHub Copilot CLI), or smaller context windows. |
| 39 | + Some Codex models and ChatGPT include deferred tools too — check your client/model |
| 40 | + directly to confirm its features so you don''t leave this enabled unnecessarily. |
| 41 | + Tools are found via ha_search_tools and executed via categorized proxies (read/write/delete). |
| 42 | + Requires restart to take effect.' |
| 43 | + enable_tool_security_policies: |
| 44 | + name: Enable Tool Security Policies (advanced) |
| 45 | + description: Gate high-stakes tool calls (lock/alarm control, automation writes, |
| 46 | + etc.) behind user approval. When a guarded tool is called, the agent tells the |
| 47 | + user to open the Tool Security Policies tab in the web UI and click Approve |
| 48 | + before the call proceeds. Per-tool rules with optional argument conditions are |
| 49 | + configured in the Tool Security Policies tab. Off by default. Requires restart |
| 50 | + to take effect. |
| 51 | + enable_security_policy_tool: |
| 52 | + name: Let AI agents edit tool security policies |
| 53 | + description: 'Registers the ha_manage_security_policy tool, which lets connected |
| 54 | + AI agents read and rewrite the tool security policies — including removing approval |
| 55 | + gates. Off by default. Independent of the Tool Security Policies option above: |
| 56 | + this controls who can edit the rules, not whether they are enforced. The tool |
| 57 | + cannot see or decide pending approvals. To keep a human in the loop, gate the |
| 58 | + tool before enabling it: switch on "security gated" for Manage Security Policy |
| 59 | + in the Tools tab of the web UI. Requires restart to take effect.' |
| 60 | + read_only_mode: |
| 61 | + name: Read Only Mode |
| 62 | + description: Toggles all write tools off, and removes ability for tools to make |
| 63 | + any write or destructive calls. Mixed read/write tools (backups, add-ons, energy |
| 64 | + preferences, voice pipelines, and code mode when enabled) stay available with |
| 65 | + their write operations blocked. Same toggle as the web UI Tools tab. Off by |
| 66 | + default. Requires restart to take effect. |
| 67 | + redact_secrets: |
| 68 | + name: Redact Secrets |
| 69 | + description: Redacts secrets from tool responses before they reach the AI assistant. |
| 70 | + Add-on options and integration fields marked as passwords are replaced with |
| 71 | + a set/empty marker (so "is this credential configured?" stays answerable), and |
| 72 | + other tool responses are scrubbed of secret values the server has already seen |
| 73 | + (values shorter than 6 characters are skipped, since replacing tiny fragments |
| 74 | + would corrupt unrelated output). Fields not marked as passwords by their schema |
| 75 | + cannot be detected. Off by default. |
| 76 | + enable_beta_features: |
| 77 | + name: Enable beta features (master) |
| 78 | + description: ⚠ DANGER — these tools can PERMANENTLY DAMAGE your Home Assistant |
| 79 | + installation. They write to your YAML config, your filesystem, install custom |
| 80 | + components, run arbitrary sandboxed Python, and edit tool docstrings the AI |
| 81 | + sees. There is no warranty and no support guarantee — you enable them at your |
| 82 | + OWN RISK. Take a Home Assistant backup before turning this on, and never enable |
| 83 | + in production without one. Master gate for the experimental sub-flags below; |
| 84 | + sub-flags are ignored at runtime while this master is off, even when explicitly |
| 85 | + set to true. The same toggle is also surfaced in the web settings UI under "Beta |
| 86 | + features (dangerous)" — either surface reflects the other on restart. |
| 87 | + enable_yaml_config_editing: |
| 88 | + name: Enable YAML config editing (beta) |
| 89 | + description: Beta feature. Allows AI assistants to add, replace, or remove top-level |
| 90 | + keys in configuration.yaml and packages/*.yaml. Only whitelisted keys are allowed |
| 91 | + (e.g., template, sensor, command_line, mqtt, knx); core keys like homeassistant, |
| 92 | + http, and recorder are blocked. Each edit validates YAML syntax, runs a config |
| 93 | + check, and creates an automatic backup. Changes to most keys require a full |
| 94 | + HA restart to take effect. See docs/beta.md for known limitations. Dedicated |
| 95 | + tools (automations, scripts, scenes, helpers, template sensors) should be preferred |
| 96 | + when available. REQUIRES the master "Enable beta features" toggle above (and |
| 97 | + in the web UI) to be on — otherwise this sub-flag is ignored at runtime regardless |
| 98 | + of its value here. |
| 99 | + enable_yaml_packages_automation: |
| 100 | + name: Allow automation in packages/*.yaml |
| 101 | + description: Sub-toggle of YAML config editing. When on, ha_config_set_yaml accepts |
| 102 | + yaml_path='automation' inside packages/*.yaml. When off, the call is rejected |
| 103 | + both client-side and server-side. The storage-mode tool (ha_config_set_automation) |
| 104 | + is unaffected. Default OFF; only takes effect when "Enable YAML config editing" |
| 105 | + above is also on. |
| 106 | + enable_yaml_packages_script: |
| 107 | + name: Allow script in packages/*.yaml |
| 108 | + description: Sub-toggle of YAML config editing. When on, ha_config_set_yaml accepts |
| 109 | + yaml_path='script' inside packages/*.yaml. When off, the call is rejected both |
| 110 | + client-side and server-side. The storage-mode tool (ha_config_set_script) is |
| 111 | + unaffected. Default OFF; only takes effect when "Enable YAML config editing" |
| 112 | + above is also on. |
| 113 | + enable_yaml_packages_scene: |
| 114 | + name: Allow scene in packages/*.yaml |
| 115 | + description: Sub-toggle of YAML config editing. When on, ha_config_set_yaml accepts |
| 116 | + yaml_path='scene' inside packages/*.yaml. When off, the call is rejected both |
| 117 | + client-side and server-side. The storage-mode tool (ha_config_set_scene) is |
| 118 | + unaffected. Default OFF; only takes effect when "Enable YAML config editing" |
| 119 | + above is also on. |
| 120 | + enable_yaml_edit_confirm: |
| 121 | + name: Require confirmation for YAML edits (diff preview) |
| 122 | + description: 'Sub-toggle of YAML config editing, ON by default (recommended). |
| 123 | + The first ha_config_set_yaml call returns a unified diff of exactly what would |
| 124 | + change on disk plus a confirm token and writes nothing; the edit only lands |
| 125 | + when the call is repeated with that token. Exists so unintended changes outside |
| 126 | + the requested edit are caught before they reach disk (issue #1720). Turn off |
| 127 | + only to save one round-trip per edit.' |
| 128 | + enable_code_mode: |
| 129 | + name: Enable custom tool sandbox (beta) |
| 130 | + description: Beta feature. Enables the ha_manage_custom_tool tool, which lets |
| 131 | + AI assistants create, run, save, and delete custom Python code in a secure sandbox |
| 132 | + when no built-in tool can handle the request. Code runs in an isolated interpreter |
| 133 | + with no filesystem or arbitrary network access. Sandbox code can hit the HA |
| 134 | + REST API (api_get/api_post), send WebSocket commands (ws_send), call existing |
| 135 | + MCP tools (call_tool), or remove a saved tool (delete_saved_tool). Saved tools |
| 136 | + persist to /data/saved_tools.json by default so they survive add-on restarts, |
| 137 | + and are visible to any client that can connect. See docs/beta.md for known limitations. |
| 138 | + Requires restart to take effect. REQUIRES the master "Enable beta features" |
| 139 | + toggle above (and in the web UI) to be on — otherwise this sub-flag is ignored |
| 140 | + at runtime regardless of its value here. |
| 141 | + enable_lite_docstrings: |
| 142 | + name: Enable lite tool docstrings (beta) |
| 143 | + description: 'Beta feature. Replaces the docstrings on a handful of heavy ha-mcp |
| 144 | + tools (automations, scripts, scenes, helpers, dashboards, ha_call_service, ha_config_set_yaml) |
| 145 | + with shorter variants that defer schema and example detail to the ha_get_skill_guide |
| 146 | + tool (or its skill:// resource). WARNING: this reduces idle token usage, but |
| 147 | + may degrade LLM performance — the trimmed descriptions rely on the LLM actually |
| 148 | + calling the skill tool or reading the skill resource for detail, which is not |
| 149 | + guaranteed (some models will skip the extra tool call and end up with less guidance |
| 150 | + than they had before). Best paired with a client that supports MCP resources |
| 151 | + or with enable_tool_search. Requires restart to take effect. REQUIRES the master |
| 152 | + "Enable beta features" toggle above (and in the web UI) to be on — otherwise |
| 153 | + this sub-flag is ignored at runtime regardless of its value here.' |
| 154 | + enable_mandatory_bps: |
| 155 | + name: Attach best-practice skills on writes |
| 156 | + description: 'Master switch for the write-tool skill content delivery feature |
| 157 | + (issue #1182). When enabled (default), the six config write tools (automations, |
| 158 | + scripts, scenes, helpers, dashboards, raw YAML) attach the canonical Home Assistant |
| 159 | + best-practice reference files under `skill_content` on every successful write, |
| 160 | + plus auto-embed any reference sections cited by best-practice warnings. Each |
| 161 | + tool also exposes a per-call `MandatoryBPS` parameter the agent can set to false |
| 162 | + on subsequent calls once it has the content. When this master switch is off, |
| 163 | + NO skill_content goes out regardless of the per-call parameter or BP warnings. |
| 164 | + Recommended ON as the first choice; disable only for models with very small |
| 165 | + context windows. Turning this off may degrade write accuracy. Requires restart |
| 166 | + to take effect.' |
| 167 | + enable_strict_mandatory_bps: |
| 168 | + name: Strict best-practices mode |
| 169 | + description: 'Strict mode: prevents the client from using the tool until it can |
| 170 | + prove that it read the best practices. While on, the six best-practice write |
| 171 | + tools (automations, scripts, scenes, helpers, dashboards, raw YAML) are blocked |
| 172 | + and return an error directing the client to read the best-practices skill via |
| 173 | + ha_get_skill_guide and pass back the acknowledgment key it obtains there. While |
| 174 | + on, the ha_get_skill_guide tool is locked enabled — it is the only publisher |
| 175 | + of the acknowledgment key. Child of the "Attach best-practice skills on writes" |
| 176 | + option above and inert while that parent is off. Requires restart to take effect.' |
| 177 | + enable_filesystem_tools: |
| 178 | + name: Enable filesystem tools (beta) |
| 179 | + description: 'Sets HAMCP_ENABLE_FILESYSTEM_TOOLS=true. Enables direct file read/write |
| 180 | + access to your Home Assistant filesystem. WARNING: This gives the MCP server |
| 181 | + sensitive direct file access to your system. Only enable if you trust the AI |
| 182 | + assistant with file operations. Requires restart to take effect. REQUIRES the |
| 183 | + master "Enable beta features" toggle above (and in the web UI) to be on — otherwise |
| 184 | + this sub-flag is ignored at runtime regardless of its value here.' |
| 185 | + enable_dashboard_screenshot: |
| 186 | + name: Enable dashboard screenshot mode (beta) |
| 187 | + description: Beta feature — disabled by default. Adds the ha_get_dashboard_screenshot |
| 188 | + tool plus include_screenshot / return_screenshot options on the dashboard get/set |
| 189 | + tools, so AI assistants can inspect one or more responsive Lovelace images (e.g. |
| 190 | + to verify a dashboard they just created). Supports stable named views, mobile/tablet/desktop |
| 191 | + batches, and PNG/JPEG/WebP/BMP output. Rendering runs in a separate, opt-in |
| 192 | + engine — balloob's "Puppet" add-on (headless Chromium) — which you install once |
| 193 | + (add balloob's add-on repository, then install "Puppet") and give a long-lived |
| 194 | + access token; on Docker/Container deployments you run that engine as a sidecar |
| 195 | + and set HAMCP_DASHBOARD_SCREENSHOT_ENGINE_URL. Nothing heavy is installed unless |
| 196 | + you both enable this and install the engine. Requires restart to take effect. |
| 197 | + REQUIRES the master "Enable beta features" toggle above (and in the web UI) |
| 198 | + to be on — otherwise this sub-flag is ignored at runtime regardless of its value |
| 199 | + here. |
| 200 | + enable_auto_backup: |
| 201 | + name: Enable auto-backup of edits |
| 202 | + description: 'Captures a per-entity snapshot before every wrapped write/destructive |
| 203 | + MCP tool call (automation, script, scene, helper, dashboard, dashboard resource, |
| 204 | + label, category, group, zone, area, calendar, todo, entity, device, integration, |
| 205 | + and sibling remove/delete tools). The auto-backup throttle bounds how often |
| 206 | + that happens: 0 (default) captures on every such call, N>0 at most one snapshot |
| 207 | + per N minutes per entity. Snapshots are saved as YAML files under /data/ha_mcp_backups/ |
| 208 | + (override via HAMCP_BACKUP_DIR) and listed, restored, or deleted via the Backups |
| 209 | + tab in the web settings UI or via ha_manage_backup(scope=''edits'', ...). For |
| 210 | + those domains the capture is best-effort — a failure logs a WARNING and the |
| 211 | + write proceeds. The file and raw-YAML tools are the exception: ha_write_file, |
| 212 | + ha_delete_file and ha_config_set_yaml treat the snapshot as a precondition, |
| 213 | + so a write through them is refused while this is off, and a failed capture blocks |
| 214 | + it. On by default; unchecking opts out of the best-effort captures and takes |
| 215 | + the file and raw-YAML tools out of service with them. Requires restart to take |
| 216 | + effect.' |
| 217 | + auto_backup_throttle_minutes: |
| 218 | + name: Auto-backup throttle (minutes) |
| 219 | + description: Per-entity throttle window. 0 (default) captures a snapshot on every |
| 220 | + wrapped write. N>0 captures at most one snapshot per N minutes per entity. Range |
| 221 | + 0–1440. |
| 222 | + auto_backup_retain_per_entity: |
| 223 | + name: Auto-backup retention (per entity) |
| 224 | + description: Maximum number of snapshots kept per entity. Older snapshots beyond |
| 225 | + this cap are rotated out on each successful capture. Default 100, range 1–10000. |
| 226 | + enable_snapshot_delete: |
| 227 | + name: Allow snapshot deletion |
| 228 | + description: Lets ha_manage_backup delete full HA snapshot tarballs (scope='snapshot', |
| 229 | + action='delete'). Off by default — a snapshot may be the last recovery point |
| 230 | + after a mistaken change, so a human must opt in here. Even when on, scheduled |
| 231 | + backups, the newest remaining snapshot, and anything younger than the age floor |
| 232 | + below stay protected. |
| 233 | + snapshot_delete_min_age_days: |
| 234 | + name: Minimum snapshot age to delete (days) |
| 235 | + description: A snapshot must be at least this old before it can be deleted. Range |
| 236 | + 0–365; 0 disables the floor (the newest-snapshot and scheduled-backup protections |
| 237 | + still apply). Default 7. |
| 238 | + tool_search_max_results: |
| 239 | + name: Tool search max results |
| 240 | + description: 'Maximum number of tools returned by ha_search_tools when tool search |
| 241 | + is enabled. Lower values (2-3) save context tokens but may miss relevant tools. |
| 242 | + Range: 2-10. Requires restart.' |
| 243 | + disabled_tools: |
| 244 | + name: Disabled tools (comma-separated) |
| 245 | + description: Tools listed here are forced off and locked in the web Settings UI |
| 246 | + (Tools tab) — they cannot be re-enabled there without unsetting this option |
| 247 | + first. A small set of mandatory tools (ha_search, ha_get_overview, ha_get_state, |
| 248 | + ha_report_issue, ha_manage_backup) cannot be disabled and keep running even |
| 249 | + if listed here. ha_get_skill_guide can be disabled only while strict best-practices |
| 250 | + mode (enable_strict_mandatory_bps) is off. Comma-separated tool names (e.g. |
| 251 | + ha_call_event,ha_eval_template). |
| 252 | + pinned_tools: |
| 253 | + name: Pinned tools (comma-separated) |
| 254 | + description: Tools listed here are locked pinned — they appear at the top of the |
| 255 | + Tools tab and cannot be unpinned from the web Settings UI without unsetting |
| 256 | + this option first. Useful in combination with Tool Search. Comma-separated tool |
| 257 | + names. |
| 258 | + verify_ssl: |
| 259 | + name: Verify TLS certificate |
| 260 | + description: Verify the Home Assistant server's TLS certificate. The add-on connects |
| 261 | + via the Supervisor proxy, so this normally has no effect and should stay enabled. |
| 262 | + Disable only if you've reconfigured the add-on to talk to HA over a public HTTPS |
| 263 | + hostname with a self-signed certificate or hostname mismatch. Disabling weakens |
| 264 | + transport security — leave on unless you know you need it. Requires restart |
| 265 | + to take effect. |
0 commit comments