config-sample.md Codex Docs, 2026-02-19 20:37 UTC → 2026-03-05 06:22 UTC

1# Sample Configuration1# Sample Configuration

2 2 

3A complete example config.toml you can copy and adapt

4 

5Use this example configuration as a starting point. It includes most keys Codex reads from `config.toml`, along with defaults and short notes.3Use this example configuration as a starting point. It includes most keys Codex reads from `config.toml`, along with defaults and short notes.

6 4 

7For explanations and guidance, see:5For explanations and guidance, see:

9- [Config basics](https://developers.openai.com/codex/config-basic)7- [Config basics](https://developers.openai.com/codex/config-basic)

10- [Advanced Config](https://developers.openai.com/codex/config-advanced)8- [Advanced Config](https://developers.openai.com/codex/config-advanced)

11- [Config Reference](https://developers.openai.com/codex/config-reference)9- [Config Reference](https://developers.openai.com/codex/config-reference)

10- [Sandbox and approvals](https://developers.openai.com/codex/security#sandbox-and-approvals)

11- [Managed configuration](https://developers.openai.com/codex/security#managed-configuration)

12 12 

13Use the snippet below as a reference. Copy only the keys and sections you need into `~/.codex/config.toml` (or into a project-scoped `.codex/config.toml`), then adjust values for your setup.13Use the snippet below as a reference. Copy only the keys and sections you need into `~/.codex/config.toml` (or into a project-scoped `.codex/config.toml`), then adjust values for your setup.

14 14 

49# model_context_window = 128000 # tokens; default: auto for model49# model_context_window = 128000 # tokens; default: auto for model

50# model_auto_compact_token_limit = 0 # tokens; unset uses model defaults50# model_auto_compact_token_limit = 0 # tokens; unset uses model defaults

51# tool_output_token_limit = 10000 # tokens stored per tool output; default: 10000 for gpt-5.2-codex51# tool_output_token_limit = 10000 # tokens stored per tool output; default: 10000 for gpt-5.2-codex

52# model_catalog_json = "/absolute/path/to/models.json" # optional startup-only model catalog override

53# background_terminal_max_timeout = 300000 # ms; max empty write_stdin poll window (default 5m)

52# log_dir = "/absolute/path/to/codex-logs" # directory for Codex logs; default: "$CODEX_HOME/log"54# log_dir = "/absolute/path/to/codex-logs" # directory for Codex logs; default: "$CODEX_HOME/log"

55# sqlite_home = "/absolute/path/to/codex-state" # optional SQLite-backed runtime state directory

53 56 

54################################################################################57################################################################################

55# Reasoning & Verbosity (Responses API capable models)58# Reasoning & Verbosity (Responses API capable models)

107# - untrusted: only known-safe read-only commands auto-run; others prompt110# - untrusted: only known-safe read-only commands auto-run; others prompt

108# - on-request: model decides when to ask (default)111# - on-request: model decides when to ask (default)

109# - never: never prompt (risky)112# - never: never prompt (risky)

113# - { reject = { ... } }: auto-reject selected prompt categories

110approval_policy = "on-request"114approval_policy = "on-request"

115# Example granular auto-reject policy:

116# approval_policy = { reject = { sandbox_approval = true, rules = false, mcp_elicitations = false } }

117 

118# Allow login-shell semantics for shell-based tools when they request `login = true`.

119# Default: true. Set false to force non-login shells and reject explicit login-shell requests.

120allow_login_shell = true

111 121 

112# Filesystem/network sandbox policy for tool calls:122# Filesystem/network sandbox policy for tool calls:

113# - read-only (default)123# - read-only (default)

115# - danger-full-access (no sandbox; extremely risky)125# - danger-full-access (no sandbox; extremely risky)

116sandbox_mode = "read-only"126sandbox_mode = "read-only"

117 127 

128[windows]

129# Native Windows sandbox mode (Windows only): unelevated | elevated

130sandbox = "unelevated"

131 

118################################################################################132################################################################################

119# Authentication & Login133# Authentication & Login

120################################################################################134################################################################################

138# Optional fixed port for MCP OAuth callback: 1-65535. Default: unset.152# Optional fixed port for MCP OAuth callback: 1-65535. Default: unset.

139# mcp_oauth_callback_port = 4321153# mcp_oauth_callback_port = 4321

140 154 

155# Optional redirect URI override for MCP OAuth login (for example, remote devbox ingress).

156# Custom callback paths are supported. `mcp_oauth_callback_port` still controls the listener port.

157# mcp_oauth_callback_url = "https://devbox.example.internal/callback"

158 

141################################################################################159################################################################################

142# Project Documentation Controls160# Project Documentation Controls

143################################################################################161################################################################################

194# Active profile name. When unset, no profile is applied.212# Active profile name. When unset, no profile is applied.

195# profile = "default"213# profile = "default"

196 214 

215################################################################################

216# Agents (multi-agent roles and limits)

217################################################################################

218 

219# [agents]

220# Maximum concurrently open agent threads. Default: 6

221# max_threads = 6

222# Maximum nested spawn depth. Root session starts at depth 0. Default: 1

223# max_depth = 1

224# Default timeout per worker for spawn_agents_on_csv jobs. When unset, the tool defaults to 1800 seconds.

225# job_max_runtime_seconds = 1800

226 

227# [agents.reviewer]

228# description = "Find security, correctness, and test risks in code."

229# config_file = "./agents/reviewer.toml" # relative to the config.toml that defines it

230 

197################################################################################231################################################################################

198# Skills (per-skill overrides)232# Skills (per-skill overrides)

199################################################################################233################################################################################

200 234 

201# Disable or re-enable a specific skill without deleting it.235# Disable or re-enable a specific skill without deleting it.

202[[skills.config]]236[[skills.config]]

203# path = "/path/to/skill"237# path = "/path/to/skill/SKILL.md"

204# enabled = false238# enabled = false

205 239 

206################################################################################240################################################################################

276# Control alternate screen usage (auto skips it in Zellij to preserve scrollback).310# Control alternate screen usage (auto skips it in Zellij to preserve scrollback).

277# alternate_screen = "auto"311# alternate_screen = "auto"

278 312 

279# Ordered list of footer status-line item IDs. Default: null (disabled).313# Ordered list of footer status-line item IDs. When unset, Codex uses:

314# ["model-with-reasoning", "context-remaining", "current-dir"].

315# Set to [] to hide the footer.

280# status_line = ["model", "context-remaining", "git-branch"]316# status_line = ["model", "context-remaining", "git-branch"]

281 317 

318# Syntax-highlighting theme (kebab-case). Use /theme in the TUI to preview and save.

319# You can also add custom .tmTheme files under $CODEX_HOME/themes.

320# theme = "catppuccin-mocha"

321 

282# Control whether users can submit feedback from `/feedback`. Default: true322# Control whether users can submit feedback from `/feedback`. Default: true

283[feedback]323[feedback]

284enabled = true324enabled = true

301 341 

302[features]342[features]

303# Leave this table empty to accept defaults. Set explicit booleans to opt in/out.343# Leave this table empty to accept defaults. Set explicit booleans to opt in/out.

304shell_tool = true344# shell_tool = true

305# apps = false345# apps = false

306# apps_mcp_gateway = false346# apps_mcp_gateway = false

307# Deprecated legacy toggles; prefer the top-level `web_search` setting.

308# web_search = false

309# web_search_cached = false347# web_search_cached = false

310# web_search_request = false348# web_search_request = false

311unified_exec = false349# unified_exec = false

312shell_snapshot = false350# shell_snapshot = false

313apply_patch_freeform = false351# apply_patch_freeform = false

352# multi_agent = false

314# search_tool = false353# search_tool = false

315# personality = true354# personality = true

316request_rule = true355# request_rule = true

317collaboration_modes = true356# collaboration_modes = true

318use_linux_sandbox_bwrap = false357# use_linux_sandbox_bwrap = false

319experimental_windows_sandbox = false358# remote_models = false

320elevated_windows_sandbox = false359# runtime_metrics = false

321remote_models = false360# powershell_utf8 = true

322runtime_metrics = false361# child_agents_md = false

323powershell_utf8 = true

324child_agents_md = false

325 362 

326################################################################################363################################################################################

327# Define MCP servers under this table. Leave empty to disable.364# Define MCP servers under this table. Leave empty to disable.

411# model_verbosity = "medium"448# model_verbosity = "medium"

412# personality = "friendly" # or "pragmatic" or "none"449# personality = "friendly" # or "pragmatic" or "none"

413# chatgpt_base_url = "https://chatgpt.com/backend-api/"450# chatgpt_base_url = "https://chatgpt.com/backend-api/"

451# model_catalog_json = "./models.json"

414# experimental_compact_prompt_file = "./compact_prompt.txt"452# experimental_compact_prompt_file = "./compact_prompt.txt"

415# include_apply_patch_tool = false453# include_apply_patch_tool = false

416# experimental_use_unified_exec_tool = false454# experimental_use_unified_exec_tool = false

424 462 

425# Optional per-app controls.463# Optional per-app controls.

426[apps]464[apps]

465# [_default] applies to all apps unless overridden per app.

466# [apps._default]

467# enabled = true

468# destructive_enabled = true

469# open_world_enabled = true

470#

427# [apps.google_drive]471# [apps.google_drive]

428# enabled = false472# enabled = false

429# disabled_reason = "user" # or "unknown"473# destructive_enabled = false # block destructive-hint tools for this app

474# default_tools_enabled = true

475# default_tools_approval_mode = "prompt" # auto | prompt | approve

476#

477# [apps.google_drive.tools."files/delete"]

478# enabled = false

479# approval_mode = "approve"

430 480 

431################################################################################481################################################################################

432# Projects (trust levels)482# Projects (trust levels)

477# client-certificate = "/etc/codex/certs/client.pem"527# client-certificate = "/etc/codex/certs/client.pem"

478# client-private-key = "/etc/codex/certs/client-key.pem"528# client-private-key = "/etc/codex/certs/client-key.pem"

479```529```

530 

531################################################################################

532 

533# Windows

534 

535################################################################################

536 

537[windows]

538 

539# Native Windows sandbox mode (Windows only): unelevated | elevated

540 

541sandbox = "unelevated"