Articles tagged with #Codex
Codex landed your work on main, on a leftover branch, or split it across two — because cloud tasks default to your repo's default branch and local worktrees collide. How to force clean per-task isolation, plus how to rescue commits that already landed wrong.
Codex hits conflict markers and stalls, picks the wrong side, or commits the markers into the file. Rebase in setup, an AGENTS.md stop-on-conflict rule, and a pre-commit hook fix it.
Codex edited package.json but never ran npm install, so package-lock.json diverges and npm ci fails. Fix it with setup.sh, a CI sync check, and AGENTS.md rules.
Codex's PR is green in CI but the app crashes after deploy. Why agent fixes that target the test surface miss the runtime, and the smoke-gate that closes the gap.
Codex defines a fresh User or ApiResponse type when an identical one already lives in your repo. Make the agent search first using AGENTS.md, a types barrel, ripgrep, and a ts-morph CI check.
Codex ran git commit --amend, an interactive rebase, or force-pushed and rewrote shared history. Lock the agent to forward-only commits with AGENTS.md, the sandbox network toggle, and a GitHub ruleset.
Codex emits a one-line PR body with no detail. Force a structured description (before/after, why, test plan) with a PR template, AGENTS.md, and a CI gate; plus the Codex Cloud button caveat.
Codex reports green tests but the failing cases were filtered out, marked .skip, or bailed early. How to force honest test reporting before merge.
Codex hits a context limit mid-task and leaves a half-applied patch. Scope tasks to one verb, split into multi-PR plans, tune auto-compaction, and resume cleanly with config that works as of June 2026.
Codex injects API patterns deprecated years ago because its training data lags. Lock it to modern APIs with AGENTS.md rules, the @typescript-eslint/no-deprecated lint rule, and current reference docs.
Side-by-side on PRDs, JIRA grooming, and doc cleanup — with current pricing, models, and a 90-minute test you can run on your own week of work.
Codex writes a file, Prettier reformats it on save, and the diff explodes. Fix it by running the project's pinned Prettier inside the agent loop and aligning AGENTS.md style hints to .prettierrc.
On 500k+ line repos Codex loses the thread halfway. Fix it by scoping the working set, pre-feeding directory summaries, and putting conventions in AGENTS.md so they survive compaction.
Codex quietly drops PNGs, PDFs, sqlite DBs, and compiled artifacts from audits. Fix it by probing binaries (file + wc -c + sha256sum), allowlisting text extensions, and forcing a no-omit output schema.
Codex re-reads the same file 8 times and re-greps the same query 5 times. Fastest fix: run /plan first, add an AGENTS.md repo map, and keep reasoning effort at medium. Full diagnosis below.
Codex halts halfway through a multi-step task with no error — usually context compaction, a sandbox/approval denial, or a premature stop signal. Diagnose by the last tool call, then resume with codex resume --last.
A verified Codex CLI keyboard shortcut reference: cancel and interrupt keys, newline with Ctrl+J, Ctrl+G external editor, Ctrl+R history search, Esc Esc to edit, Shift+Tab approval modes, and terminal fixes.
How Codex Agent Skills work in June 2026: what a SKILL.md is, the exact .agents/skills/ load order, how /skills and $name invocation differ, and skill vs AGENTS.md vs MCP vs plugin.
Codex shows "unable to clone" or "Repository not found"? Usually the ChatGPT Codex Connector GitHub App isn't installed on the org, the repo isn't in its allowlist, or your SSO/OAuth grant lapsed. Fix it in priority order.
Codex stalls at Setting up environment — usually a runtime mismatch, a private-registry token, or an install that blows past the cached container. The fast fixes, verified for June 2026.
50 files, 1,500 lines, nobody will review it. Cap diff size with Codex Plan Mode, an AGENTS.md PR budget, and a CI gate — and split the one you already have.
How OpenAI Codex runs sandboxed cloud and CLI coding tasks in 2026: which ChatGPT plan you need, the GPT-5.5 default, the task-spec format that produces reviewable PRs, and the mistakes that bite first-time users.
Set up OpenAI Codex as a pre-reviewer with the /review command, @codex review on GitHub, and AGENTS.md guidelines that produce P0/P1 findings.
Use Codex CLI to spot-check sitemap correctness, freshness, and hreflang coverage against your real build output.
Asked Codex to audit your project and got a 50-bullet report mixing typos with architecture? Re-scope to one dimension, cap output, force file:line, use /review's custom-instructions preset.
Codex says "done" but leaves broken imports, half-converted types, and untouched call sites. Bind "done" to a verifier in AGENTS.md, then check the exit code yourself.
Codex creates `utils.v2.ts` next to `utils.ts` instead of editing in place. The real cause is an apply_patch Update→Add fallback. Force in-place edits with AGENTS.md rules, git, and the right sandbox mode.
Codex skips the build, misreads it, or trusts a truncated tail. Fix it with exit-code checks, machine-readable verifiers, and a Stop hook gate — not prose summaries.
The reported bug is gone, two adjacent features regressed. Cap blast radius, enumerate call sites, prefer caller-side guards over shared-util changes.
New files land in /src/ when your repo uses /app/; deps go to the root package.json in a monorepo; tests use a framework you don't have. Fix it with a root AGENTS.md, nested overrides, and canonical-example pointers — verified against Codex behavior, June 2026.
Codex uses a column, env var, or function signature it never read. Force a quote-before-write rule and run Codex read-only so it grounds every line in your real files.
Codex writes `getUserById` when your codebase uses `findUserById`. Codify each convention as rule + a canonical example file in AGENTS.md, then verify it loaded.
Codex fails with "Invalid Context" or "No such file or directory" when applying a patch. Tell the two apart, then refresh state or restart the session.
Codex review returns generic bullets like "consider error handling." Fix it with the built-in /review command, AGENTS.md review guidelines, and file:line-anchored prompts.
Functionally correct but reads like a stranger wrote it: async/await mixed with .then(), wrong import order, off comment style. Anchor Codex to a canonical file in AGENTS.md, then enforce the rest with lint.
"Test the happy path and error path" is useless filler. Force Codex to bind tests to the real function signature, the actual types, edge cases, and bug history.
Codex puts files in the wrong place or ignores your conventions. Constrain it with AGENTS.md, a sibling-file reference, and plan-first prompting.
Keep coding agents on-spec across a multi-file build: a CLAUDE.md/AGENTS.md template, a 3-part per-task prompt, and the spec-file rules that matter as of June 2026.
Run a technical SEO review with Codex using copy-paste prompts for canonical, hreflang, title/meta, sitemap, and JSON-LD checks — plus shell commands that verify every finding.
OpenAI Codex and Anthropic Claude Code are both agentic coding tools, but they run in different places and bill differently. A specific, up-to-date comparison with prices, models, and benchmarks.