Answer: Add admin-enforced command rules in the `[rules]` section using `prefix_rules` with `decision = "prompt"` for sensitive commands, and enforce approval settings with `allowed_approval_policies = ["on-request", "untrusted"]` and `allowed_approvals_reviewers = ["user"]`. Status: Confirmed Entry: tool/codex/permissions-and-security/require-approval-for-sensitive-commands Revision: 2 Digest: sha256:b14217315ae3b776 (changes whenever the content changes) Timeline: Created: 2026-11-13T03:50:25Z Last changed: 2026-11-20T13:38:36Z (forbidden decision: — → forbidden) Last verified: 2026-11-20T13:38:36Z Next check: 2026-11-21T13:38:36Z Check interval: Daily Expected change: No fixed schedule Data: Field | Value | Source | As of rule configuration | prefix_rules | [1] | 2026-11-20 prompt decision | prompt | [1] | 2026-11-20 allowed approval policies | ["on-request", "untrusted"] | [1] | 2026-11-20 allowed approvals reviewers | ["user"] | [1] | 2026-11-20 forbidden decision | forbidden | [1] | 2026-11-20 Sources: [1] OpenAI. "Codex security cookbook". https://learn.chatgpt.com/docs/enterprise/security-cookbook.md Role: primary. Accessed: 2026-11-20T13:38:36Z Quote: "Add admin-enforced command rules: ```toml [rules] prefix_rules" (Supports: rule configuration) Quote: "{ pattern = [{ token = "git" }, { any_of = ["push", "commit"] }], decision = "prompt", justification = "Require review before mutating shared Git state." }," (Supports: prompt decision) Quote: "allowed_approval_policies = ["on-request", "untrusted"]" (Supports: allowed approval policies) Quote: "allowed_approvals_reviewers = ["user"]" (Supports: allowed approvals reviewers) Quote: "{ pattern = [{ token = "rm" }], decision = "forbidden", justification = "Destructive removal is not permitted in managed Codex workflows." }," (Supports: forbidden decision) Snapshot SHA-256: bd59a6b1e051756cc5174ba6dad0c24b937039ee3450b8f48a48a4c4563db12e (captured 2026-11-19T13:29:01Z) Verification: Method: Automated extraction from [1] with deepseek-flash (JSON output, thinking disabled). Every value was checked by code to appear verbatim in a quoted passage of [1]. On this check the text of [1] was unchanged since 2026-11-19T13:29:03Z, so the values verified then were kept. Cross-check: not-available at 2026-11-20T13:38:36Z. No independent second source is configured for this entry Last successful check: 2026-11-20T13:38:36Z Last attempt: 2026-11-20T13:38:36Z Notes and limitations: Extracted automatically from a single official source (section "Require approval for sensitive commands"). Related: https://latestn.com/tool/codex/permissions-and-security/restrict-network-destinations How do I restrict network destinations for Codex Cloud and local Codex? https://latestn.com/tool/codex/permissions-and-security/review-findings-across-scans How do I review findings across scans in the Codex Security workbench? https://latestn.com/tool/codex/permissions-and-security/review-sensitive-actions How do I review sensitive Codex actions before they run? https://latestn.com/tool/codex/permissions-and-security/preset-domain-lists What preset domain lists are available for Codex Cloud (Legacy) internet access? https://latestn.com/tool/codex/permissions-and-security/prisma-airs-enforcement-modes What enforcement modes and failure behaviors can I choose for Prisma AIRS in Codex? https://latestn.com/tool/codex/permissions-and-security/read-only-triage-requirements What are the requirements to run read-only triage in Codex? https://latestn.com/tool/ All entries in AI tools Change log: Time | Event | Change | Source 2026-11-20T13:38:36Z | changed | forbidden decision: — → forbidden | [1] 2026-11-20T13:38:36Z | changed | prompt decision: — → prompt | [1] 2026-11-20T13:38:36Z | changed | rule configuration: — → prefix_rules | [1] 2026-11-20T13:38:36Z | changed | destructive host commands decision: forbidden → — | [1] 2026-11-20T13:38:36Z | changed | kubectl rule decision: prompt → — | [1] 2026-11-20T13:38:36Z | changed | terraform rule decision: prompt → — | [1] 2026-11-20T13:38:36Z | changed | git rule decision: prompt → — | [1] 2026-11-13T03:50:25Z | added | Entry created | — Changes for this entry as a feed: https://latestn.com/tool/codex/permissions-and-security/require-approval-for-sensitive-commands/latest Challenge this entry: Open challenges: None How: If a value here is wrong or out of date, send what is wrong and a link to evidence. Challenges are queued and checked against the evidence; they do not change the entry by themselves. If the entry changes as a result, the change log says so. Form: https://latestn.com/dispute/tool/codex/permissions-and-security/require-approval-for-sensitive-commands POST: https://latestn.com/dispute/tool/codex/permissions-and-security/require-approval-for-sensitive-commands with claim, evidence_url, and optionally field and contact (form-encoded or JSON) GET: https://latestn.com/dispute/tool/codex/permissions-and-security/require-approval-for-sensitive-commands?claim={what is wrong}&evidence_url={link} How to cite: LatestN. (2026). How do I require approval for sensitive commands in Codex? Revision 2, last verified 2026-11-20T13:38:36Z. https://latestn.com/tool/codex/permissions-and-security/require-approval-for-sensitive-commands (accessed 2026-11-21T10:03:22Z). @misc{LatestN_tool_codex_permissions_and_security_require_approval_for_sensitive_commands_r2, title = {How do I require approval for sensitive commands in Codex?}, author = {{LatestN}}, year = {2026}, howpublished = {\url{https://latestn.com/tool/codex/permissions-and-security/require-approval-for-sensitive-commands}}, note = {Revision 2; digest sha256:b14217315ae3b776; last verified 2026-11-20T13:38:36Z; accessed 2026-11-21T10:03:22Z} } Reuse: License: CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/) Attribution: Copying and adapting are allowed, including commercially, if you credit LatestN and link to the entry you used. Each entry ends with a ready-made citation. URL: https://latestn.com/tool/codex/permissions-and-security/require-approval-for-sensitive-commands