config-reference.md Codex Docs, 2026-03-07 18:10 UTC → 2026-05-05 23:00 UTC

8 8 

9For sandbox and approval keys (`approval_policy`, `sandbox_mode`, and `sandbox_workspace_write.*`), pair this reference with [Sandbox and approvals](https://developers.openai.com/codex/agent-approvals-security#sandbox-and-approvals), [Protected paths in writable roots](https://developers.openai.com/codex/agent-approvals-security#protected-paths-in-writable-roots), and [Network access](https://developers.openai.com/codex/agent-approvals-security#network-access).9For sandbox and approval keys (`approval_policy`, `sandbox_mode`, and `sandbox_workspace_write.*`), pair this reference with [Sandbox and approvals](https://developers.openai.com/codex/agent-approvals-security#sandbox-and-approvals), [Protected paths in writable roots](https://developers.openai.com/codex/agent-approvals-security#protected-paths-in-writable-roots), and [Network access](https://developers.openai.com/codex/agent-approvals-security#network-access).

10 10 

11| Key | Type / Values | Details |11<ConfigTable

12| --- | --- | --- |12 options={[

13| `agents.<name>.config_file` | `string (path)` | Path to a TOML config layer for that role; relative paths resolve from the config file that declares the role. |13 {

14| `agents.<name>.description` | `string` | Role guidance shown to Codex when choosing and spawning that agent type. |14 key: "model",

15| `agents.job_max_runtime_seconds` | `number` | Default per-worker timeout for `spawn_agents_on_csv` jobs. When unset, the tool falls back to 1800 seconds per worker. |15 type: "string",

16| `agents.max_depth` | `number` | Maximum nesting depth allowed for spawned agent threads (root sessions start at depth 0; default: 1). |16 description: "Model to use (e.g., `gpt-5.5`).",

17| `agents.max_threads` | `number` | Maximum number of agent threads that can be open concurrently. |17 },

18| `allow_login_shell` | `boolean` | Allow shell-based tools to use login-shell semantics. Defaults to `true`; when `false`, `login = true` requests are rejected and omitted `login` defaults to non-login shells. |18 {

19| `approval_policy` | `untrusted | on-request | never | { reject = { sandbox_approval = bool, rules = bool, mcp_elicitations = bool } }` | Controls when Codex pauses for approval before executing commands. You can also use `approval_policy = { reject = { ... } }` to auto-reject specific prompt categories while keeping other prompts interactive. `on-failure` is deprecated; use `on-request` for interactive runs or `never` for non-interactive runs. |19 key: "review_model",

20| `approval_policy.reject.mcp_elicitations` | `boolean` | When `true`, MCP elicitation prompts are auto-rejected instead of shown to the user. |20 type: "string",

21| `approval_policy.reject.rules` | `boolean` | When `true`, approvals triggered by execpolicy `prompt` rules are auto-rejected. |21 description:

22| `approval_policy.reject.sandbox_approval` | `boolean` | When `true`, sandbox escalation approval prompts are auto-rejected. |22 "Optional model override used by `/review` (defaults to the current session model).",

23| `apps._default.destructive_enabled` | `boolean` | Default allow/deny for app tools with `destructive_hint = true`. |23 },

24| `apps._default.enabled` | `boolean` | Default app enabled state for all apps unless overridden per app. |24 {

25| `apps._default.open_world_enabled` | `boolean` | Default allow/deny for app tools with `open_world_hint = true`. |25 key: "model_provider",

26| `apps.<id>.default_tools_approval_mode` | `auto | prompt | approve` | Default approval behavior for tools in this app unless a per-tool override exists. |26 type: "string",

27| `apps.<id>.default_tools_enabled` | `boolean` | Default enabled state for tools in this app unless a per-tool override exists. |27 description: "Provider id from `model_providers` (default: `openai`).",

28| `apps.<id>.destructive_enabled` | `boolean` | Allow or block tools in this app that advertise `destructive_hint = true`. |28 },

29| `apps.<id>.enabled` | `boolean` | Enable or disable a specific app/connector by id (default: true). |29 {

30| `apps.<id>.open_world_enabled` | `boolean` | Allow or block tools in this app that advertise `open_world_hint = true`. |30 key: "openai_base_url",

31| `apps.<id>.tools.<tool>.approval_mode` | `auto | prompt | approve` | Per-tool approval behavior override for a single app tool. |31 type: "string",

32| `apps.<id>.tools.<tool>.enabled` | `boolean` | Per-tool enabled override for an app tool (for example `repos/list`). |32 description:

33| `background_terminal_max_timeout` | `number` | Maximum poll window in milliseconds for empty `write_stdin` polls (background terminal polling). Default: `300000` (5 minutes). Replaces the older `background_terminal_timeout` key. |33 "Base URL override for the built-in `openai` model provider.",

34| `chatgpt_base_url` | `string` | Override the base URL used during the ChatGPT login flow. |34 },

35| `check_for_update_on_startup` | `boolean` | Check for Codex updates on startup (set to false only when updates are centrally managed). |35 {

36| `cli_auth_credentials_store` | `file | keyring | auto` | Control where the CLI stores cached credentials (file-based auth.json vs OS keychain). |36 key: "model_context_window",

37| `compact_prompt` | `string` | Inline override for the history compaction prompt. |37 type: "number",

38| `developer_instructions` | `string` | Additional developer instructions injected into the session (optional). |38 description: "Context window tokens available to the active model.",

39| `disable_paste_burst` | `boolean` | Disable burst-paste detection in the TUI. |39 },

40| `experimental_compact_prompt_file` | `string (path)` | Load the compaction prompt override from a file (experimental). |40 {

41| `experimental_use_freeform_apply_patch` | `boolean` | Legacy name for enabling freeform apply\_patch; prefer `[features].apply_patch_freeform` or `codex --enable apply_patch_freeform`. |41 key: "model_auto_compact_token_limit",

42| `experimental_use_unified_exec_tool` | `boolean` | Legacy name for enabling unified exec; prefer `[features].unified_exec` or `codex --enable unified_exec`. |42 type: "number",

43| `features.apply_patch_freeform` | `boolean` | Expose the freeform `apply_patch` tool (experimental). |43 description:

44| `features.apps` | `boolean` | Enable ChatGPT Apps/connectors support (experimental). |44 "Token threshold that triggers automatic history compaction (unset uses model defaults).",

45| `features.apps_mcp_gateway` | `boolean` | Route Apps MCP calls through the OpenAI connectors MCP gateway (`https://api.openai.com/v1/connectors/mcp/`) instead of legacy routing (experimental). |45 },

46| `features.child_agents_md` | `boolean` | Append AGENTS.md scope/precedence guidance even when no AGENTS.md is present (experimental). |46 {

47| `features.collaboration_modes` | `boolean` | Enable collaboration modes such as plan mode (stable; on by default). |47 key: "model_catalog_json",

48| `features.multi_agent` | `boolean` | Enable multi-agent collaboration tools (`spawn_agent`, `send_input`, `resume_agent`, `wait`, `close_agent`, and `spawn_agents_on_csv`) (experimental; off by default). |48 type: "string (path)",

49| `features.personality` | `boolean` | Enable personality selection controls (stable; on by default). |49 description:

50| `features.powershell_utf8` | `boolean` | Force PowerShell UTF-8 output (defaults to true). |50 "Optional path to a JSON model catalog loaded on startup. Profile-level `profiles.<name>.model_catalog_json` can override this per profile.",

51| `features.remote_models` | `boolean` | Refresh remote model list before showing readiness (experimental). |51 },

52| `features.request_rule` | `boolean` | Enable Smart approvals (`prefix_rule` suggestions on escalation requests; stable; on by default). |52 {

53| `features.runtime_metrics` | `boolean` | Show runtime metrics summary in TUI turn separators (experimental). |53 key: "oss_provider",

54| `features.search_tool` | `boolean` | Enable `search_tool_bm25` for Apps tool discovery before invoking app MCP tools (experimental). |54 type: "lmstudio | ollama",

55| `features.shell_snapshot` | `boolean` | Snapshot shell environment to speed up repeated commands (beta). |55 description:

56| `features.shell_tool` | `boolean` | Enable the default `shell` tool for running commands (stable; on by default). |56 "Default local provider used when running with `--oss` (defaults to prompting if unset).",

57| `features.unified_exec` | `boolean` | Use the unified PTY-backed exec tool (beta). |57 },

58| `features.use_linux_sandbox_bwrap` | `boolean` | Use the bubblewrap-based Linux sandbox pipeline (experimental; off by default). |58 {

59| `features.web_search` | `boolean` | Deprecated legacy toggle; prefer the top-level `web_search` setting. |59 key: "approval_policy",

60| `features.web_search_cached` | `boolean` | Deprecated legacy toggle. When `web_search` is unset, true maps to `web_search = "cached"`. |60 type: "untrusted | on-request | never | { granular = { sandbox_approval = bool, rules = bool, mcp_elicitations = bool, request_permissions = bool, skill_approval = bool } }",

61| `features.web_search_request` | `boolean` | Deprecated legacy toggle. When `web_search` is unset, true maps to `web_search = "live"`. |61 description:

62| `feedback.enabled` | `boolean` | Enable feedback submission via `/feedback` across Codex surfaces (default: true). |62 "Controls when Codex pauses for approval before executing commands. You can also use `approval_policy = { granular = { ... } }` to allow or auto-reject specific prompt categories while keeping other prompts interactive. `on-failure` is deprecated; use `on-request` for interactive runs or `never` for non-interactive runs.",

63| `file_opener` | `vscode | vscode-insiders | windsurf | cursor | none` | URI scheme used to open citations from Codex output (default: `vscode`). |63 },

64| `forced_chatgpt_workspace_id` | `string (uuid)` | Limit ChatGPT logins to a specific workspace identifier. |64 {

65| `forced_login_method` | `chatgpt | api` | Restrict Codex to a specific authentication method. |65 key: "approval_policy.granular.sandbox_approval",

66| `hide_agent_reasoning` | `boolean` | Suppress reasoning events in both the TUI and `codex exec` output. |66 type: "boolean",

67| `history.max_bytes` | `number` | If set, caps the history file size in bytes by dropping oldest entries. |67 description:

68| `history.persistence` | `save-all | none` | Control whether Codex saves session transcripts to history.jsonl. |68 "When `true`, sandbox escalation approval prompts are allowed to surface.",

69| `include_apply_patch_tool` | `boolean` | Legacy name for enabling freeform apply\_patch; prefer `[features].apply_patch_freeform`. |69 },

70| `instructions` | `string` | Reserved for future use; prefer `model_instructions_file` or `AGENTS.md`. |70 {

71| `log_dir` | `string (path)` | Directory where Codex writes log files (for example `codex-tui.log`); defaults to `$CODEX_HOME/log`. |71 key: "approval_policy.granular.rules",

72| `mcp_oauth_callback_port` | `integer` | Optional fixed port for the local HTTP callback server used during MCP OAuth login. When unset, Codex binds to an ephemeral port chosen by the OS. |72 type: "boolean",

73| `mcp_oauth_callback_url` | `string` | Optional redirect URI override for MCP OAuth login (for example, a devbox ingress URL). `mcp_oauth_callback_port` still controls the callback listener port. |73 description:

74| `mcp_oauth_credentials_store` | `auto | file | keyring` | Preferred store for MCP OAuth credentials. |74 "When `true`, approvals triggered by execpolicy `prompt` rules are allowed to surface.",

75| `mcp_servers.<id>.args` | `array<string>` | Arguments passed to the MCP stdio server command. |75 },

76| `mcp_servers.<id>.bearer_token_env_var` | `string` | Environment variable sourcing the bearer token for an MCP HTTP server. |76 {

77| `mcp_servers.<id>.command` | `string` | Launcher command for an MCP stdio server. |77 key: "approval_policy.granular.mcp_elicitations",

78| `mcp_servers.<id>.cwd` | `string` | Working directory for the MCP stdio server process. |78 type: "boolean",

79| `mcp_servers.<id>.disabled_tools` | `array<string>` | Deny list applied after `enabled_tools` for the MCP server. |79 description:

80| `mcp_servers.<id>.enabled` | `boolean` | Disable an MCP server without removing its configuration. |80 "When `true`, MCP elicitation prompts are allowed to surface instead of being auto-rejected.",

81| `mcp_servers.<id>.enabled_tools` | `array<string>` | Allow list of tool names exposed by the MCP server. |81 },

82| `mcp_servers.<id>.env` | `map<string,string>` | Environment variables forwarded to the MCP stdio server. |82 {

83| `mcp_servers.<id>.env_http_headers` | `map<string,string>` | HTTP headers populated from environment variables for an MCP HTTP server. |83 key: "approval_policy.granular.request_permissions",

84| `mcp_servers.<id>.env_vars` | `array<string>` | Additional environment variables to whitelist for an MCP stdio server. |84 type: "boolean",

85| `mcp_servers.<id>.http_headers` | `map<string,string>` | Static HTTP headers included with each MCP HTTP request. |85 description:

86| `mcp_servers.<id>.required` | `boolean` | When true, fail startup/resume if this enabled MCP server cannot initialize. |86 "When `true`, prompts from the `request_permissions` tool are allowed to surface.",

87| `mcp_servers.<id>.startup_timeout_ms` | `number` | Alias for `startup_timeout_sec` in milliseconds. |87 },

88| `mcp_servers.<id>.startup_timeout_sec` | `number` | Override the default 10s startup timeout for an MCP server. |88 {

89| `mcp_servers.<id>.tool_timeout_sec` | `number` | Override the default 60s per-tool timeout for an MCP server. |89 key: "approval_policy.granular.skill_approval",

90| `mcp_servers.<id>.url` | `string` | Endpoint for an MCP streamable HTTP server. |90 type: "boolean",

91| `model` | `string` | Model to use (e.g., `gpt-5-codex`). |91 description:

92| `model_auto_compact_token_limit` | `number` | Token threshold that triggers automatic history compaction (unset uses model defaults). |92 "When `true`, skill-script approval prompts are allowed to surface.",

93| `model_catalog_json` | `string (path)` | Optional path to a JSON model catalog loaded on startup. Profile-level `profiles.<name>.model_catalog_json` can override this per profile. |93 },

94| `model_context_window` | `number` | Context window tokens available to the active model. |94 {

95| `model_instructions_file` | `string (path)` | Replacement for built-in instructions instead of `AGENTS.md`. |95 key: "approvals_reviewer",

96| `model_provider` | `string` | Provider id from `model_providers` (default: `openai`). |96 type: "user | auto_review",

97| `model_providers.<id>.base_url` | `string` | API base URL for the model provider. |97 description:

98| `model_providers.<id>.env_http_headers` | `map<string,string>` | HTTP headers populated from environment variables when present. |98 "Who reviews eligible approval prompts under `on-request` or granular approval policies. Defaults to `user`; `auto_review` uses the reviewer subagent. This setting doesn't change sandboxing or review actions already allowed inside the sandbox.",

99| `model_providers.<id>.env_key` | `string` | Environment variable supplying the provider API key. |99 },

100| `model_providers.<id>.env_key_instructions` | `string` | Optional setup guidance for the provider API key. |100 {

101| `model_providers.<id>.experimental_bearer_token` | `string` | Direct bearer token for the provider (discouraged; use `env_key`). |101 key: "auto_review.policy",

102| `model_providers.<id>.http_headers` | `map<string,string>` | Static HTTP headers added to provider requests. |102 type: "string",

103| `model_providers.<id>.name` | `string` | Display name for a custom model provider. |103 description:

104| `model_providers.<id>.query_params` | `map<string,string>` | Extra query parameters appended to provider requests. |104 "Local Markdown policy instructions for automatic review. Managed `guardian_policy_config` takes precedence. Blank values are ignored.",

105| `model_providers.<id>.request_max_retries` | `number` | Retry count for HTTP requests to the provider (default: 4). |105 },

106| `model_providers.<id>.requires_openai_auth` | `boolean` | The provider uses OpenAI authentication (defaults to false). |106 {

107| `model_providers.<id>.stream_idle_timeout_ms` | `number` | Idle timeout for SSE streams in milliseconds (default: 300000). |107 key: "allow_login_shell",

108| `model_providers.<id>.stream_max_retries` | `number` | Retry count for SSE streaming interruptions (default: 5). |108 type: "boolean",

109| `model_providers.<id>.wire_api` | `chat | responses` | Protocol used by the provider (defaults to `chat` if omitted). |109 description:

110| `model_reasoning_effort` | `minimal | low | medium | high | xhigh` | Adjust reasoning effort for supported models (Responses API only; `xhigh` is model-dependent). |110 "Allow shell-based tools to use login-shell semantics. Defaults to `true`; when `false`, `login = true` requests are rejected and omitted `login` defaults to non-login shells.",

111| `model_reasoning_summary` | `auto | concise | detailed | none` | Select reasoning summary detail or disable summaries entirely. |111 },

112| `model_supports_reasoning_summaries` | `boolean` | Force Codex to send or not send reasoning metadata. |112 {

113| `model_verbosity` | `low | medium | high` | Control GPT-5 Responses API verbosity (defaults to `medium`). |113 key: "sandbox_mode",

114| `notice.hide_full_access_warning` | `boolean` | Track acknowledgement of the full access warning prompt. |114 type: "read-only | workspace-write | danger-full-access",

115| `notice.hide_gpt-5.1-codex-max_migration_prompt` | `boolean` | Track acknowledgement of the gpt-5.1-codex-max migration prompt. |115 description:

116| `notice.hide_gpt5_1_migration_prompt` | `boolean` | Track acknowledgement of the GPT-5.1 migration prompt. |116 "Sandbox policy for filesystem and network access during command execution.",

117| `notice.hide_rate_limit_model_nudge` | `boolean` | Track opt-out of the rate limit model switch reminder. |117 },

118| `notice.hide_world_writable_warning` | `boolean` | Track acknowledgement of the Windows world-writable directories warning. |118 {

119| `notice.model_migrations` | `map<string,string>` | Track acknowledged model migrations as old->new mappings. |119 key: "sandbox_workspace_write.writable_roots",

120| `notify` | `array<string>` | Command invoked for notifications; receives a JSON payload from Codex. |120 type: "array<string>",

121| `oss_provider` | `lmstudio | ollama` | Default local provider used when running with `--oss` (defaults to prompting if unset). |121 description:

122| `otel.environment` | `string` | Environment tag applied to emitted OpenTelemetry events (default: `dev`). |122 'Additional writable roots when `sandbox_mode = "workspace-write"`.',

123| `otel.exporter` | `none | otlp-http | otlp-grpc` | Select the OpenTelemetry exporter and provide any endpoint metadata. |123 },

124| `otel.exporter.<id>.endpoint` | `string` | Exporter endpoint for OTEL logs. |124 {

125| `otel.exporter.<id>.headers` | `map<string,string>` | Static headers included with OTEL exporter requests. |125 key: "sandbox_workspace_write.network_access",

126| `otel.exporter.<id>.protocol` | `binary | json` | Protocol used by the OTLP/HTTP exporter. |126 type: "boolean",

127| `otel.exporter.<id>.tls.ca-certificate` | `string` | CA certificate path for OTEL exporter TLS. |127 description:

128| `otel.exporter.<id>.tls.client-certificate` | `string` | Client certificate path for OTEL exporter TLS. |128 "Allow outbound network access inside the workspace-write sandbox.",

129| `otel.exporter.<id>.tls.client-private-key` | `string` | Client private key path for OTEL exporter TLS. |129 },

130| `otel.log_user_prompt` | `boolean` | Opt in to exporting raw user prompts with OpenTelemetry logs. |130 {

131| `otel.trace_exporter` | `none | otlp-http | otlp-grpc` | Select the OpenTelemetry trace exporter and provide any endpoint metadata. |131 key: "sandbox_workspace_write.exclude_tmpdir_env_var",

132| `otel.trace_exporter.<id>.endpoint` | `string` | Trace exporter endpoint for OTEL logs. |132 type: "boolean",

133| `otel.trace_exporter.<id>.headers` | `map<string,string>` | Static headers included with OTEL trace exporter requests. |133 description:

134| `otel.trace_exporter.<id>.protocol` | `binary | json` | Protocol used by the OTLP/HTTP trace exporter. |134 "Exclude `$TMPDIR` from writable roots in workspace-write mode.",

135| `otel.trace_exporter.<id>.tls.ca-certificate` | `string` | CA certificate path for OTEL trace exporter TLS. |135 },

136| `otel.trace_exporter.<id>.tls.client-certificate` | `string` | Client certificate path for OTEL trace exporter TLS. |136 {

137| `otel.trace_exporter.<id>.tls.client-private-key` | `string` | Client private key path for OTEL trace exporter TLS. |137 key: "sandbox_workspace_write.exclude_slash_tmp",

138| `personality` | `none | friendly | pragmatic` | Default communication style for models that advertise `supportsPersonality`; can be overridden per thread/turn or via `/personality`. |138 type: "boolean",

139| `profile` | `string` | Default profile applied at startup (equivalent to `--profile`). |139 description:

140| `profiles.<name>.*` | `various` | Profile-scoped overrides for any of the supported configuration keys. |140 "Exclude `/tmp` from writable roots in workspace-write mode.",

141| `profiles.<name>.experimental_use_freeform_apply_patch` | `boolean` | Legacy name for enabling freeform apply\_patch; prefer `[features].apply_patch_freeform`. |141 },

142| `profiles.<name>.experimental_use_unified_exec_tool` | `boolean` | Legacy name for enabling unified exec; prefer `[features].unified_exec`. |142 {

143| `profiles.<name>.include_apply_patch_tool` | `boolean` | Legacy name for enabling freeform apply\_patch; prefer `[features].apply_patch_freeform`. |143 key: "windows.sandbox",

144| `profiles.<name>.model_catalog_json` | `string (path)` | Profile-scoped model catalog JSON path override (applied on startup only; overrides the top-level `model_catalog_json` for that profile). |144 type: "unelevated | elevated",

145| `profiles.<name>.oss_provider` | `lmstudio | ollama` | Profile-scoped OSS provider for `--oss` sessions. |145 description:

146| `profiles.<name>.personality` | `none | friendly | pragmatic` | Profile-scoped communication style override for supported models. |146 "Windows-only native sandbox mode when running Codex natively on Windows.",

147| `profiles.<name>.web_search` | `disabled | cached | live` | Profile-scoped web search mode override (default: `"cached"`). |147 },

148| `project_doc_fallback_filenames` | `array<string>` | Additional filenames to try when `AGENTS.md` is missing. |148 {

149| `project_doc_max_bytes` | `number` | Maximum bytes read from `AGENTS.md` when building project instructions. |149 key: "windows.sandbox_private_desktop",

150| `project_root_markers` | `array<string>` | List of project root marker filenames; used when searching parent directories for the project root. |150 type: "boolean",

151| `projects.<path>.trust_level` | `string` | Mark a project or worktree as trusted or untrusted (`"trusted"` | `"untrusted"`). Untrusted projects skip project-scoped `.codex/` layers. |151 description:

152| `review_model` | `string` | Optional model override used by `/review` (defaults to the current session model). |152 "Run the final sandboxed child process on a private desktop by default on native Windows. Set `false` only for compatibility with the older `Winsta0\\\\Default` behavior.",

153| `sandbox_mode` | `read-only | workspace-write | danger-full-access` | Sandbox policy for filesystem and network access during command execution. |153 },

154| `sandbox_workspace_write.exclude_slash_tmp` | `boolean` | Exclude `/tmp` from writable roots in workspace-write mode. |154 {

155| `sandbox_workspace_write.exclude_tmpdir_env_var` | `boolean` | Exclude `$TMPDIR` from writable roots in workspace-write mode. |155 key: "notify",

156| `sandbox_workspace_write.network_access` | `boolean` | Allow outbound network access inside the workspace-write sandbox. |156 type: "array<string>",

157| `sandbox_workspace_write.writable_roots` | `array<string>` | Additional writable roots when `sandbox_mode = "workspace-write"`. |157 description:

158| `shell_environment_policy.exclude` | `array<string>` | Glob patterns for removing environment variables after the defaults. |158 "Command invoked for notifications; receives a JSON payload from Codex.",

159| `shell_environment_policy.experimental_use_profile` | `boolean` | Use the user shell profile when spawning subprocesses. |159 },

160| `shell_environment_policy.ignore_default_excludes` | `boolean` | Keep variables containing KEY/SECRET/TOKEN before other filters run. |160 {

161| `shell_environment_policy.include_only` | `array<string>` | Whitelist of patterns; when set only matching variables are kept. |161 key: "check_for_update_on_startup",

162| `shell_environment_policy.inherit` | `all | core | none` | Baseline environment inheritance when spawning subprocesses. |162 type: "boolean",

163| `shell_environment_policy.set` | `map<string,string>` | Explicit environment overrides injected into every subprocess. |163 description:

164| `show_raw_agent_reasoning` | `boolean` | Surface raw reasoning content when the active model emits it. |164 "Check for Codex updates on startup (set to false only when updates are centrally managed).",

165| `skills.config` | `array<object>` | Per-skill enablement overrides stored in config.toml. |165 },

166| `skills.config.<index>.enabled` | `boolean` | Enable or disable the referenced skill. |166 {

167| `skills.config.<index>.path` | `string (path)` | Path to a skill folder containing `SKILL.md`. |167 key: "feedback.enabled",

168| `sqlite_home` | `string (path)` | Directory where Codex stores the SQLite-backed state DB used by agent jobs and other resumable runtime state. |168 type: "boolean",

169| `suppress_unstable_features_warning` | `boolean` | Suppress the warning that appears when under-development feature flags are enabled. |169 description:

170| `tool_output_token_limit` | `number` | Token budget for storing individual tool/function outputs in history. |170 "Enable feedback submission via `/feedback` across Codex surfaces (default: true).",

171| `tools.web_search` | `boolean` | Deprecated legacy toggle for web search; prefer the top-level `web_search` setting. |171 },

172| `tui` | `table` | TUI-specific options such as enabling inline desktop notifications. |172 {

173| `tui.alternate_screen` | `auto | always | never` | Control alternate screen usage for the TUI (default: auto; auto skips it in Zellij to preserve scrollback). |173 key: "analytics.enabled",

174| `tui.animations` | `boolean` | Enable terminal animations (welcome screen, shimmer, spinner) (default: true). |174 type: "boolean",

175| `tui.notification_method` | `auto | osc9 | bel` | Notification method for unfocused terminal notifications (default: auto). |175 description:

176| `tui.notifications` | `boolean | array<string>` | Enable TUI notifications; optionally restrict to specific event types. |176 "Enable or disable analytics for this machine/profile. When unset, the client default applies.",

177| `tui.show_tooltips` | `boolean` | Show onboarding tooltips in the TUI welcome screen (default: true). |177 },

178| `tui.status_line` | `array<string> | null` | Ordered list of TUI footer status-line item identifiers. `null` disables the status line. |178 {

179| `web_search` | `disabled | cached | live` | Web search mode (default: `"cached"`; cached uses an OpenAI-maintained index and does not fetch live pages; if you use `--yolo` or another full access sandbox setting, it defaults to `"live"`). Use `"live"` to fetch the most recent data from the web, or `"disabled"` to remove the tool. |179 key: "instructions",

180| `windows_wsl_setup_acknowledged` | `boolean` | Track Windows onboarding acknowledgement (Windows only). |180 type: "string",

181| `windows.sandbox` | `unelevated | elevated` | Windows-only native sandbox mode when running Codex natively on Windows. |181 description:

182 182 "Reserved for future use; prefer `model_instructions_file` or `AGENTS.md`.",

183Key183 },

184 184 {

185`agents.<name>.config_file`185 key: "developer_instructions",

186 186 type: "string",

187Type / Values187 description:

188 188 "Additional developer instructions injected into the session (optional).",

189`string (path)`189 },

190 190 {

191Details191 key: "log_dir",

192 192 type: "string (path)",

193Path to a TOML config layer for that role; relative paths resolve from the config file that declares the role.193 description:

194 194 "Directory where Codex writes log files (for example `codex-tui.log`); defaults to `$CODEX_HOME/log`.",

195Key195 },

196 196 {

197`agents.<name>.description`197 key: "sqlite_home",

198 198 type: "string (path)",

199Type / Values199 description:

200 200 "Directory where Codex stores the SQLite-backed state DB used by agent jobs and other resumable runtime state.",

201`string`201 },

202 202 {

203Details203 key: "compact_prompt",

204 204 type: "string",

205Role guidance shown to Codex when choosing and spawning that agent type.205 description: "Inline override for the history compaction prompt.",

206 206 },

207Key207 {

208 208 key: "commit_attribution",

209`agents.job_max_runtime_seconds`209 type: "string",

210 210 description:

211Type / Values211 "Override the commit co-author trailer text. Set an empty string to disable automatic attribution.",

212 212 },

213`number`213 {

214 214 key: "model_instructions_file",

215Details215 type: "string (path)",

216 216 description:

217Default per-worker timeout for `spawn_agents_on_csv` jobs. When unset, the tool falls back to 1800 seconds per worker.217 "Replacement for built-in instructions instead of `AGENTS.md`.",

218 218 },

219Key219 {

220 220 key: "personality",

221`agents.max_depth`221 type: "none | friendly | pragmatic",

222 222 description:

223Type / Values223 "Default communication style for models that advertise `supportsPersonality`; can be overridden per thread/turn or via `/personality`.",

224 224 },

225`number`225 {

226 226 key: "service_tier",

227Details227 type: "flex | fast",

228 228 description: "Preferred service tier for new turns.",

229Maximum nesting depth allowed for spawned agent threads (root sessions start at depth 0; default: 1).229 },

230 230 {

231Key231 key: "experimental_compact_prompt_file",

232 232 type: "string (path)",

233`agents.max_threads`233 description:

234 234 "Load the compaction prompt override from a file (experimental).",

235Type / Values235 },

236 236 {

237`number`237 key: "skills.config",

238 238 type: "array<object>",

239Details239 description: "Per-skill enablement overrides stored in config.toml.",

240 240 },

241Maximum number of agent threads that can be open concurrently.241 {

242 242 key: "skills.config.<index>.path",

243Key243 type: "string (path)",

244 244 description: "Path to a skill folder containing `SKILL.md`.",

245`allow_login_shell`245 },

246 246 {

247Type / Values247 key: "skills.config.<index>.enabled",

248 248 type: "boolean",

249`boolean`249 description: "Enable or disable the referenced skill.",

250 250 },

251Details251 {

252 252 key: "apps.<id>.enabled",

253Allow shell-based tools to use login-shell semantics. Defaults to `true`; when `false`, `login = true` requests are rejected and omitted `login` defaults to non-login shells.253 type: "boolean",

254 254 description:

255Key255 "Enable or disable a specific app/connector by id (default: true).",

256 256 },

257`approval_policy`257 {

258 258 key: "apps._default.enabled",

259Type / Values259 type: "boolean",

260 260 description:

261`untrusted | on-request | never | { reject = { sandbox_approval = bool, rules = bool, mcp_elicitations = bool } }`261 "Default app enabled state for all apps unless overridden per app.",

262 262 },

263Details263 {

264 264 key: "apps._default.destructive_enabled",

265Controls when Codex pauses for approval before executing commands. You can also use `approval_policy = { reject = { ... } }` to auto-reject specific prompt categories while keeping other prompts interactive. `on-failure` is deprecated; use `on-request` for interactive runs or `never` for non-interactive runs.265 type: "boolean",

266 266 description:

267Key267 "Default allow/deny for app tools with `destructive_hint = true`.",

268 268 },

269`approval_policy.reject.mcp_elicitations`269 {

270 270 key: "apps._default.open_world_enabled",

271Type / Values271 type: "boolean",

272 272 description:

273`boolean`273 "Default allow/deny for app tools with `open_world_hint = true`.",

274 274 },

275Details275 {

276 276 key: "apps.<id>.destructive_enabled",

277When `true`, MCP elicitation prompts are auto-rejected instead of shown to the user.277 type: "boolean",

278 278 description:

279Key279 "Allow or block tools in this app that advertise `destructive_hint = true`.",

280 280 },

281`approval_policy.reject.rules`281 {

282 282 key: "apps.<id>.open_world_enabled",

283Type / Values283 type: "boolean",

284 284 description:

285`boolean`285 "Allow or block tools in this app that advertise `open_world_hint = true`.",

286 286 },

287Details287 {

288 288 key: "apps.<id>.default_tools_enabled",

289When `true`, approvals triggered by execpolicy `prompt` rules are auto-rejected.289 type: "boolean",

290 290 description:

291Key291 "Default enabled state for tools in this app unless a per-tool override exists.",

292 292 },

293`approval_policy.reject.sandbox_approval`293 {

294 294 key: "apps.<id>.default_tools_approval_mode",

295Type / Values295 type: "auto | prompt | approve",

296 296 description:

297`boolean`297 "Default approval behavior for tools in this app unless a per-tool override exists.",

298 298 },

299Details299 {

300 300 key: "apps.<id>.tools.<tool>.enabled",

301When `true`, sandbox escalation approval prompts are auto-rejected.301 type: "boolean",

302 302 description:

303Key303 "Per-tool enabled override for an app tool (for example `repos/list`).",

304 304 },

305`apps._default.destructive_enabled`305 {

306 306 key: "apps.<id>.tools.<tool>.approval_mode",

307Type / Values307 type: "auto | prompt | approve",

308 308 description: "Per-tool approval behavior override for a single app tool.",

309`boolean`309 },

310 310 {

311Details311 key: "tool_suggest.discoverables",

312 312 type: "array<table>",

313Default allow/deny for app tools with `destructive_hint = true`.313 description:

314 314 'Allow tool suggestions for additional discoverable connectors or plugins. Each entry uses `type = "connector"` or `"plugin"` and an `id`.',

315Key315 },

316 316 {

317`apps._default.enabled`317 key: "tool_suggest.disabled_tools",

318 318 type: "array<table>",

319Type / Values319 description:

320 320 'Disable suggestions for specific discoverable connectors or plugins. Each entry uses `type = "connector"` or `"plugin"` and an `id`.',

321`boolean`321 },

322 322 {

323Details323 key: "features.apps",

324 324 type: "boolean",

325Default app enabled state for all apps unless overridden per app.325 description: "Enable ChatGPT Apps/connectors support (experimental).",

326 326 },

327Key327 {

328 328 key: "features.codex_hooks",

329`apps._default.open_world_enabled`329 type: "boolean",

330 330 description:

331Type / Values331 "Enable lifecycle hooks loaded from `hooks.json` or inline `[hooks]` config.",

332 332 },

333`boolean`333 {

334 334 key: "hooks",

335Details335 type: "table",

336 336 description:

337Default allow/deny for app tools with `open_world_hint = true`.337 "Lifecycle hooks configured inline in `config.toml`. Uses the same event schema as `hooks.json`; see the Hooks guide for examples and supported events.",

338 338 },

339Key339 {

340 340 key: "features.memories",

341`apps.<id>.default_tools_approval_mode`341 type: "boolean",

342 342 description: "Enable [Memories](https://developers.openai.com/codex/memories) (off by default).",

343Type / Values343 },

344 344 {

345`auto | prompt | approve`345 key: "mcp_servers.<id>.command",

346 346 type: "string",

347Details347 description: "Launcher command for an MCP stdio server.",

348 348 },

349Default approval behavior for tools in this app unless a per-tool override exists.349 {

350 350 key: "mcp_servers.<id>.args",

351Key351 type: "array<string>",

352 352 description: "Arguments passed to the MCP stdio server command.",

353`apps.<id>.default_tools_enabled`353 },

354 354 {

355Type / Values355 key: "mcp_servers.<id>.env",

356 356 type: "map<string,string>",

357`boolean`357 description: "Environment variables forwarded to the MCP stdio server.",

358 358 },

359Details359 {

360 360 key: "mcp_servers.<id>.env_vars",

361Default enabled state for tools in this app unless a per-tool override exists.361 type: 'array<string | { name = string, source = "local" | "remote" }>',

362 362 description:

363Key363 'Additional environment variables to whitelist for an MCP stdio server. String entries default to `source = "local"`; use `source = "remote"` only with executor-backed remote stdio.',

364 364 },

365`apps.<id>.destructive_enabled`365 {

366 366 key: "mcp_servers.<id>.cwd",

367Type / Values367 type: "string",

368 368 description: "Working directory for the MCP stdio server process.",

369`boolean`369 },

370 370 {

371Details371 key: "mcp_servers.<id>.url",

372 372 type: "string",

373Allow or block tools in this app that advertise `destructive_hint = true`.373 description: "Endpoint for an MCP streamable HTTP server.",

374 374 },

375Key375 {

376 376 key: "mcp_servers.<id>.bearer_token_env_var",

377`apps.<id>.enabled`377 type: "string",

378 378 description:

379Type / Values379 "Environment variable sourcing the bearer token for an MCP HTTP server.",

380 380 },

381`boolean`381 {

382 382 key: "mcp_servers.<id>.http_headers",

383Details383 type: "map<string,string>",

384 384 description: "Static HTTP headers included with each MCP HTTP request.",

385Enable or disable a specific app/connector by id (default: true).385 },

386 386 {

387Key387 key: "mcp_servers.<id>.env_http_headers",

388 388 type: "map<string,string>",

389`apps.<id>.open_world_enabled`389 description:

390 390 "HTTP headers populated from environment variables for an MCP HTTP server.",

391Type / Values391 },

392 392 {

393`boolean`393 key: "mcp_servers.<id>.enabled",

394 394 type: "boolean",

395Details395 description: "Disable an MCP server without removing its configuration.",

396 396 },

397Allow or block tools in this app that advertise `open_world_hint = true`.397 {

398 398 key: "mcp_servers.<id>.required",

399Key399 type: "boolean",

400 400 description:

401`apps.<id>.tools.<tool>.approval_mode`401 "When true, fail startup/resume if this enabled MCP server cannot initialize.",

402 402 },

403Type / Values403 {

404 404 key: "mcp_servers.<id>.startup_timeout_sec",

405`auto | prompt | approve`405 type: "number",

406 406 description:

407Details407 "Override the default 10s startup timeout for an MCP server.",

408 408 },

409Per-tool approval behavior override for a single app tool.409 {

410 410 key: "mcp_servers.<id>.startup_timeout_ms",

411Key411 type: "number",

412 412 description: "Alias for `startup_timeout_sec` in milliseconds.",

413`apps.<id>.tools.<tool>.enabled`413 },

414 414 {

415Type / Values415 key: "mcp_servers.<id>.tool_timeout_sec",

416 416 type: "number",

417`boolean`417 description:

418 418 "Override the default 60s per-tool timeout for an MCP server.",

419Details419 },

420 420 {

421Per-tool enabled override for an app tool (for example `repos/list`).421 key: "mcp_servers.<id>.enabled_tools",

422 422 type: "array<string>",

423Key423 description: "Allow list of tool names exposed by the MCP server.",

424 424 },

425`background_terminal_max_timeout`425 {

426 426 key: "mcp_servers.<id>.disabled_tools",

427Type / Values427 type: "array<string>",

428 428 description:

429`number`429 "Deny list applied after `enabled_tools` for the MCP server.",

430 430 },

431Details431 {

432 432 key: "mcp_servers.<id>.scopes",

433Maximum poll window in milliseconds for empty `write_stdin` polls (background terminal polling). Default: `300000` (5 minutes). Replaces the older `background_terminal_timeout` key.433 type: "array<string>",

434 434 description:

435Key435 "OAuth scopes to request when authenticating to that MCP server.",

436 436 },

437`chatgpt_base_url`437 {

438 438 key: "mcp_servers.<id>.oauth_resource",

439Type / Values439 type: "string",

440 440 description:

441`string`441 "Optional RFC 8707 OAuth resource parameter to include during MCP login.",

442 442 },

443Details443 {

444 444 key: "mcp_servers.<id>.experimental_environment",

445Override the base URL used during the ChatGPT login flow.445 type: "local | remote",

446 446 description:

447Key447 "Experimental placement for an MCP server. `remote` starts stdio servers through a remote executor environment; streamable HTTP remote placement is not implemented.",

448 448 },

449`check_for_update_on_startup`449 {

450 450 key: "agents.max_threads",

451Type / Values451 type: "number",

452 452 description:

453`boolean`453 "Maximum number of agent threads that can be open concurrently. Defaults to `6` when unset.",

454 454 },

455Details455 {

456 456 key: "agents.max_depth",

457Check for Codex updates on startup (set to false only when updates are centrally managed).457 type: "number",

458 458 description:

459Key459 "Maximum nesting depth allowed for spawned agent threads (root sessions start at depth 0; default: 1).",

460 460 },

461`cli_auth_credentials_store`461 {

462 462 key: "agents.job_max_runtime_seconds",

463Type / Values463 type: "number",

464 464 description:

465`file | keyring | auto`465 "Default per-worker timeout for `spawn_agents_on_csv` jobs. When unset, the tool falls back to 1800 seconds per worker.",

466 466 },

467Details467 {

468 468 key: "agents.<name>.description",

469Control where the CLI stores cached credentials (file-based auth.json vs OS keychain).469 type: "string",

470 470 description:

471Key471 "Role guidance shown to Codex when choosing and spawning that agent type.",

472 472 },

473`compact_prompt`473 {

474 474 key: "agents.<name>.config_file",

475Type / Values475 type: "string (path)",

476 476 description:

477`string`477 "Path to a TOML config layer for that role; relative paths resolve from the config file that declares the role.",

478 478 },

479Details479 {

480 480 key: "agents.<name>.nickname_candidates",

481Inline override for the history compaction prompt.481 type: "array<string>",

482 482 description:

483Key483 "Optional pool of display nicknames for spawned agents in that role.",

484 484 },

485`developer_instructions`485 {

486 486 key: "memories.generate_memories",

487Type / Values487 type: "boolean",

488 488 description:

489`string`489 "When `false`, newly created threads are not stored as memory-generation inputs. Defaults to `true`.",

490 490 },

491Details491 {

492 492 key: "memories.use_memories",

493Additional developer instructions injected into the session (optional).493 type: "boolean",

494 494 description:

495Key495 "When `false`, Codex skips injecting existing memories into future sessions. Defaults to `true`.",

496 496 },

497`disable_paste_burst`497 {

498 498 key: "memories.disable_on_external_context",

499Type / Values499 type: "boolean",

500 500 description:

501`boolean`501 "When `true`, threads that use external context such as MCP tool calls, web search, or tool search are kept out of memory generation. Defaults to `false`. Legacy alias: `memories.no_memories_if_mcp_or_web_search`.",

502 502 },

503Details503 {

504 504 key: "memories.max_raw_memories_for_consolidation",

505Disable burst-paste detection in the TUI.505 type: "number",

506 506 description:

507Key507 "Maximum recent raw memories retained for global consolidation. Defaults to `256` and is capped at `4096`.",

508 508 },

509`experimental_compact_prompt_file`509 {

510 510 key: "memories.max_unused_days",

511Type / Values511 type: "number",

512 512 description:

513`string (path)`513 "Maximum days since a memory was last used before it becomes ineligible for consolidation. Defaults to `30` and is clamped to `0`-`365`.",

514 514 },

515Details515 {

516 516 key: "memories.max_rollout_age_days",

517Load the compaction prompt override from a file (experimental).517 type: "number",

518 518 description:

519Key519 "Maximum age of threads considered for memory generation. Defaults to `30` and is clamped to `0`-`90`.",

520 520 },

521`experimental_use_freeform_apply_patch`521 {

522 522 key: "memories.max_rollouts_per_startup",

523Type / Values523 type: "number",

524 524 description:

525`boolean`525 "Maximum rollout candidates processed per startup pass. Defaults to `16` and is capped at `128`.",

526 526 },

527Details527 {

528 528 key: "memories.min_rollout_idle_hours",

529Legacy name for enabling freeform apply\_patch; prefer `[features].apply_patch_freeform` or `codex --enable apply_patch_freeform`.529 type: "number",

530 530 description:

531Key531 "Minimum idle time before a thread is considered for memory generation. Defaults to `6` and is clamped to `1`-`48`.",

532 532 },

533`experimental_use_unified_exec_tool`533 {

534 534 key: "memories.min_rate_limit_remaining_percent",

535Type / Values535 type: "number",

536 536 description:

537`boolean`537 "Minimum remaining percentage required in Codex rate-limit windows before memory generation starts. Defaults to `25` and is clamped to `0`-`100`.",

538 538 },

539Details539 {

540 540 key: "memories.extract_model",

541Legacy name for enabling unified exec; prefer `[features].unified_exec` or `codex --enable unified_exec`.541 type: "string",

542 542 description: "Optional model override for per-thread memory extraction.",

543Key543 },

544 544 {

545`features.apply_patch_freeform`545 key: "memories.consolidation_model",

546 546 type: "string",

547Type / Values547 description: "Optional model override for global memory consolidation.",

548 548 },

549`boolean`549 {

550 550 key: "features.unified_exec",

551Details551 type: "boolean",

552 552 description:

553Expose the freeform `apply_patch` tool (experimental).553 "Use the unified PTY-backed exec tool (stable; enabled by default except on Windows).",

554 554 },

555Key555 {

556 556 key: "features.shell_snapshot",

557`features.apps`557 type: "boolean",

558 558 description:

559Type / Values559 "Snapshot shell environment to speed up repeated commands (stable; on by default).",

560 560 },

561`boolean`561 {

562 562 key: "features.undo",

563Details563 type: "boolean",

564 564 description: "Enable undo support (stable; off by default).",

565Enable ChatGPT Apps/connectors support (experimental).565 },

566 566 {

567Key567 key: "features.multi_agent",

568 568 type: "boolean",

569`features.apps_mcp_gateway`569 description:

570 570 "Enable multi-agent collaboration tools (`spawn_agent`, `send_input`, `resume_agent`, `wait_agent`, and `close_agent`) (stable; on by default).",

571Type / Values571 },

572 572 {

573`boolean`573 key: "features.personality",

574 574 type: "boolean",

575Details575 description:

576 576 "Enable personality selection controls (stable; on by default).",

577Route Apps MCP calls through the OpenAI connectors MCP gateway (`https://api.openai.com/v1/connectors/mcp/`) instead of legacy routing (experimental).577 },

578 578 {

579Key579 key: "features.web_search",

580 580 type: "boolean",

581`features.child_agents_md`581 description:

582 582 "Deprecated legacy toggle; prefer the top-level `web_search` setting.",

583Type / Values583 },

584 584 {

585`boolean`585 key: "features.web_search_cached",

586 586 type: "boolean",

587Details587 description:

588 588 'Deprecated legacy toggle. When `web_search` is unset, true maps to `web_search = "cached"`.',

589Append AGENTS.md scope/precedence guidance even when no AGENTS.md is present (experimental).589 },

590 590 {

591Key591 key: "features.web_search_request",

592 592 type: "boolean",

593`features.collaboration_modes`593 description:

594 594 'Deprecated legacy toggle. When `web_search` is unset, true maps to `web_search = "live"`.',

595Type / Values595 },

596 596 {

597`boolean`597 key: "features.shell_tool",

598 598 type: "boolean",

599Details599 description:

600 600 "Enable the default `shell` tool for running commands (stable; on by default).",

601Enable collaboration modes such as plan mode (stable; on by default).601 },

602 602 {

603Key603 key: "features.enable_request_compression",

604 604 type: "boolean",

605`features.multi_agent`605 description:

606 606 "Compress streaming request bodies with zstd when supported (stable; on by default).",

607Type / Values607 },

608 608 {

609`boolean`609 key: "features.skill_mcp_dependency_install",

610 610 type: "boolean",

611Details611 description:

612 612 "Allow prompting and installing missing MCP dependencies for skills (stable; on by default).",

613Enable multi-agent collaboration tools (`spawn_agent`, `send_input`, `resume_agent`, `wait`, `close_agent`, and `spawn_agents_on_csv`) (experimental; off by default).613 },

614 614 {

615Key615 key: "features.fast_mode",

616 616 type: "boolean",

617`features.personality`617 description:

618 618 'Enable Fast mode selection and the `service_tier = "fast"` path (stable; on by default).',

619Type / Values619 },

620 620 {

621`boolean`621 key: "features.prevent_idle_sleep",

622 622 type: "boolean",

623Details623 description:

624 624 "Prevent the machine from sleeping while a turn is actively running (experimental; off by default).",

625Enable personality selection controls (stable; on by default).625 },

626 626 {

627Key627 key: "suppress_unstable_features_warning",

628 628 type: "boolean",

629`features.powershell_utf8`629 description:

630 630 "Suppress the warning that appears when under-development feature flags are enabled.",

631Type / Values631 },

632 632 {

633`boolean`633 key: "model_providers.<id>",

634 634 type: "table",

635Details635 description:

636 636 "Custom provider definition. Built-in provider IDs (`openai`, `ollama`, and `lmstudio`) are reserved and cannot be overridden.",

637Force PowerShell UTF-8 output (defaults to true).637 },

638 638 {

639Key639 key: "model_providers.<id>.name",

640 640 type: "string",

641`features.remote_models`641 description: "Display name for a custom model provider.",

642 642 },

643Type / Values643 {

644 644 key: "model_providers.<id>.base_url",

645`boolean`645 type: "string",

646 646 description: "API base URL for the model provider.",

647Details647 },

648 648 {

649Refresh remote model list before showing readiness (experimental).649 key: "model_providers.<id>.env_key",

650 650 type: "string",

651Key651 description: "Environment variable supplying the provider API key.",

652 652 },

653`features.request_rule`653 {

654 654 key: "model_providers.<id>.env_key_instructions",

655Type / Values655 type: "string",

656 656 description: "Optional setup guidance for the provider API key.",

657`boolean`657 },

658 658 {

659Details659 key: "model_providers.<id>.experimental_bearer_token",

660 660 type: "string",

661Enable Smart approvals (`prefix_rule` suggestions on escalation requests; stable; on by default).661 description:

662 662 "Direct bearer token for the provider (discouraged; use `env_key`).",

663Key663 },

664 664 {

665`features.runtime_metrics`665 key: "model_providers.<id>.requires_openai_auth",

666 666 type: "boolean",

667Type / Values667 description:

668 668 "The provider uses OpenAI authentication (defaults to false).",

669`boolean`669 },

670 670 {

671Details671 key: "model_providers.<id>.wire_api",

672 672 type: "responses",

673Show runtime metrics summary in TUI turn separators (experimental).673 description:

674 674 "Protocol used by the provider. `responses` is the only supported value, and it is the default when omitted.",

675Key675 },

676 676 {

677`features.search_tool`677 key: "model_providers.<id>.query_params",

678 678 type: "map<string,string>",

679Type / Values679 description: "Extra query parameters appended to provider requests.",

680 680 },

681`boolean`681 {

682 682 key: "model_providers.<id>.http_headers",

683Details683 type: "map<string,string>",

684 684 description: "Static HTTP headers added to provider requests.",

685Enable `search_tool_bm25` for Apps tool discovery before invoking app MCP tools (experimental).685 },

686 686 {

687Key687 key: "model_providers.<id>.env_http_headers",

688 688 type: "map<string,string>",

689`features.shell_snapshot`689 description:

690 690 "HTTP headers populated from environment variables when present.",

691Type / Values691 },

692 692 {

693`boolean`693 key: "model_providers.<id>.request_max_retries",

694 694 type: "number",

695Details695 description:

696 696 "Retry count for HTTP requests to the provider (default: 4).",

697Snapshot shell environment to speed up repeated commands (beta).697 },

698 698 {

699Key699 key: "model_providers.<id>.stream_max_retries",

700 700 type: "number",

701`features.shell_tool`701 description: "Retry count for SSE streaming interruptions (default: 5).",

702 702 },

703Type / Values703 {

704 704 key: "model_providers.<id>.stream_idle_timeout_ms",

705`boolean`705 type: "number",

706 706 description:

707Details707 "Idle timeout for SSE streams in milliseconds (default: 300000).",

708 708 },

709Enable the default `shell` tool for running commands (stable; on by default).709 {

710 710 key: "model_providers.<id>.supports_websockets",

711Key711 type: "boolean",

712 712 description:

713`features.unified_exec`713 "Whether that provider supports the Responses API WebSocket transport.",

714 714 },

715Type / Values715 {

716 716 key: "model_providers.<id>.auth",

717`boolean`717 type: "table",

718 718 description:

719Details719 "Command-backed bearer token configuration for a custom provider. Do not combine with `env_key`, `experimental_bearer_token`, or `requires_openai_auth`.",

720 720 },

721Use the unified PTY-backed exec tool (beta).721 {

722 722 key: "model_providers.<id>.auth.command",

723Key723 type: "string",

724 724 description:

725`features.use_linux_sandbox_bwrap`725 "Command to run when Codex needs a bearer token. The command must print the token to stdout.",

726 726 },

727Type / Values727 {

728 728 key: "model_providers.<id>.auth.args",

729`boolean`729 type: "array<string>",

730 730 description: "Arguments passed to the token command.",

731Details731 },

732 732 {

733Use the bubblewrap-based Linux sandbox pipeline (experimental; off by default).733 key: "model_providers.<id>.auth.timeout_ms",

734 734 type: "number",

735Key735 description:

736 736 "Maximum token command runtime in milliseconds (default: 5000).",

737`features.web_search`737 },

738 738 {

739Type / Values739 key: "model_providers.<id>.auth.refresh_interval_ms",

740 740 type: "number",

741`boolean`741 description:

742 742 "How often Codex proactively refreshes the token in milliseconds (default: 300000). Set to `0` to refresh only after an authentication retry.",

743Details743 },

744 744 {

745Deprecated legacy toggle; prefer the top-level `web_search` setting.745 key: "model_providers.<id>.auth.cwd",

746 746 type: "string (path)",

747Key747 description: "Working directory for the token command.",

748 748 },

749`features.web_search_cached`749 {

750 750 key: "model_providers.amazon-bedrock.aws.profile",

751Type / Values751 type: "string",

752 752 description:

753`boolean`753 "AWS profile name used by the built-in `amazon-bedrock` provider.",

754 754 },

755Details755 {

756 756 key: "model_providers.amazon-bedrock.aws.region",

757Deprecated legacy toggle. When `web_search` is unset, true maps to `web_search = "cached"`.757 type: "string",

758 758 description: "AWS region used by the built-in `amazon-bedrock` provider.",

759Key759 },

760 760 {

761`features.web_search_request`761 key: "model_reasoning_effort",

762 762 type: "minimal | low | medium | high | xhigh",

763Type / Values763 description:

764 764 "Adjust reasoning effort for supported models (Responses API only; `xhigh` is model-dependent).",

765`boolean`765 },

766 766 {

767Details767 key: "plan_mode_reasoning_effort",

768 768 type: "none | minimal | low | medium | high | xhigh",

769Deprecated legacy toggle. When `web_search` is unset, true maps to `web_search = "live"`.769 description:

770 770 "Plan-mode-specific reasoning override. When unset, Plan mode uses its built-in preset default.",

771Key771 },

772 772 {

773`feedback.enabled`773 key: "model_reasoning_summary",

774 774 type: "auto | concise | detailed | none",

775Type / Values775 description:

776 776 "Select reasoning summary detail or disable summaries entirely.",

777`boolean`777 },

778 778 {

779Details779 key: "model_verbosity",

780 780 type: "low | medium | high",

781Enable feedback submission via `/feedback` across Codex surfaces (default: true).781 description:

782 782 "Optional GPT-5 Responses API verbosity override; when unset, the selected model/preset default is used.",

783Key783 },

784 784 {

785`file_opener`785 key: "model_supports_reasoning_summaries",

786 786 type: "boolean",

787Type / Values787 description: "Force Codex to send or not send reasoning metadata.",

788 788 },

789`vscode | vscode-insiders | windsurf | cursor | none`789 {

790 790 key: "shell_environment_policy.inherit",

791Details791 type: "all | core | none",

792 792 description:

793URI scheme used to open citations from Codex output (default: `vscode`).793 "Baseline environment inheritance when spawning subprocesses.",

794 794 },

795Key795 {

796 796 key: "shell_environment_policy.ignore_default_excludes",

797`forced_chatgpt_workspace_id`797 type: "boolean",

798 798 description:

799Type / Values799 "Keep variables containing KEY/SECRET/TOKEN before other filters run.",

800 800 },

801`string (uuid)`801 {

802 802 key: "shell_environment_policy.exclude",

803Details803 type: "array<string>",

804 804 description:

805Limit ChatGPT logins to a specific workspace identifier.805 "Glob patterns for removing environment variables after the defaults.",

806 806 },

807Key807 {

808 808 key: "shell_environment_policy.include_only",

809`forced_login_method`809 type: "array<string>",

810 810 description:

811Type / Values811 "Whitelist of patterns; when set only matching variables are kept.",

812 812 },

813`chatgpt | api`813 {

814 814 key: "shell_environment_policy.set",

815Details815 type: "map<string,string>",

816 816 description:

817Restrict Codex to a specific authentication method.817 "Explicit environment overrides injected into every subprocess.",

818 818 },

819Key819 {

820 820 key: "shell_environment_policy.experimental_use_profile",

821`hide_agent_reasoning`821 type: "boolean",

822 822 description: "Use the user shell profile when spawning subprocesses.",

823Type / Values823 },

824 824 {

825`boolean`825 key: "project_root_markers",

826 826 type: "array<string>",

827Details827 description:

828 828 "List of project root marker filenames; used when searching parent directories for the project root.",

829Suppress reasoning events in both the TUI and `codex exec` output.829 },

830 830 {

831Key831 key: "project_doc_max_bytes",

832 832 type: "number",

833`history.max_bytes`833 description:

834 834 "Maximum bytes read from `AGENTS.md` when building project instructions.",

835Type / Values835 },