AI Tool Profile
Codex
Codex is OpenAI's coding agent for writing, reviewing, and shipping code across local projects, editors, terminals, cloud tasks, and automated workflows.
Before you choose · Build a useful app
Codex: workflow and access check
Build and test the beginner calculator or another small scoped change.
Review due · Official-source review; no new hands-on test
This review covers the notes below. Earlier prices, tests and Product Facts retain their original dates.
OpenAI documents Codex CLI inspecting a local repository, editing files and running local tools. The guide recommends Git checkpoints before and after a task. Codex CLI documentation
Check before choosing: Confirm sign-in, permissions and available usage; build a preview and record actual tests. The earlier pricing table keeps its September 8 scope.
Return to Build a useful app → · Browse retained reviews and tests
Review scope and source notes
Earlier article text, prices, tests and Product Facts keep their original dates and scope. This check covers only the cited notes here; it does not refresh the whole profile or its machine-readable facts.
Checked by: Codex-assisted official-source review. Task fit is editorial guidance. Test the exact plan, model and route with your own fixture.
Commercial relationship: Not established by this source review. Existing article disclosures still apply.
- Codex CLI documentation — Inspect, edit, and run code from your terminal; Getting started. Checked 2026-09-20.
Maintenance history
- 2026-09-20 · source recheck · Source recheck
Coverage describes this profile, not a product score or a current price guarantee.
Kingy AI Product Facts
Codex
Current statusUnknown
- Company
- Not yet reviewed
- Primary job
- Inspect code, make changes, run commands, and automate repeatable work without leaving your terminal.
- Audience
- Not yet reviewed
- Free plan
- Available
- Evidence
- Source-backed; review due
- Latest source check
- October 20, 2026
See all tracked product factsPricing, platforms, dependencies, data claims, regions, and timeline
Sources and checks
- Coverage
- 8 of 87 tracked fields currently verified
- Sources
- 6
Source-check dates apply to the tracked facts. Editorial updates elsewhere on this page may have a different date.
Plans and pricing units
- Plus Price not listed in this summary; check the official pricing source. Consumer account
- Free Price not listed in this summary; check the official pricing source. Consumer account
- Business Price not listed in this summary; check the official pricing source. Business account
- Go Price not listed in this summary; check the official pricing source. Consumer account
- API Key Price not listed in this summary; check the official pricing source. Api account
- Enterprise & Edu Price not listed in this summary; check the official pricing source. Enterprise account
- Pro Price not listed in this summary; check the official pricing source. Consumer account
Where it runs
- Cli
- AvailableOfficial sources: Codex CLI official identity evidence · CLI – Codex | OpenAI Developers
- Product surface
- Available
- Product surface
- Available
- Product surface — Install or access
- https://marketplace.visualstudio.com/items?itemName=openai.chatgptSource: Codex IDE extension
- Product surface
- Available
- Product surface
- Available
Stack and integrations
- Model / provider dependencies
- Not yet reviewed
- Integrations
- Not yet reviewed
Data and regions
- Vendor data-use claim
- Not yet reviewed
- Vendor retention claim
- Not yet reviewed
- Regions
- Not yet reviewed
Launch and latest material update
- Launch
- Not yet reviewed
- Latest material update
- Not yet reviewed
Review history
- Codex — Verification refreshed
Kingy reconfirmed the published facts against the cited sources.
View published revision - Codex — Verification refreshed
Kingy reconfirmed the published facts against the cited sources.
View published revision - Codex — Verification refreshed
Kingy reconfirmed the published facts against the cited sources.
View published revision - Codex — Verification refreshed
Kingy reconfirmed the published facts against the cited sources.
View published revision - Codex — Verification refreshed
Kingy reconfirmed the published facts against the cited sources.
View published revision - Codex — Verification refreshed
Kingy reconfirmed the published facts against the cited sources.
View published revision - Codex — Correction published
Kingy published an approved correction to this Product Facts record.
View published revision - Codex — Product Facts updated
Kingy published an approved change to this Product Facts record.
View published revision
Current purchase and access routes
Pricing rechecked October 8, 2026: OpenAI’s pricing page now explicitly lists Pro plans at $100, $200 or $500 USD per month. That currency statement applies to Pro. The other quoted plan displays do not establish an ISO currency, universal regional availability, taxes or a checkout total. Access, security and historical statements retain the review dates stated below.
Answer first: OpenAI’s current Codex pricing page displays Free at $0/month, Go at $8/month, Plus at $20/month and Pro from $100/month. The Free, Go and Plus cards do not name an ISO currency or universal region, so Kingy preserves their dollar symbols. Pro explicitly lists $100, $200 and $500 USD monthly tiers. Business is displayed at $20 per user per month with two or more users when billed annually, or $25 per user per month when billed monthly; that card likewise does not name an ISO currency or universal region. Enterprise and Edu require sales contact.
| Route | Captured display | Access boundary |
|---|---|---|
| Free | $0/month | For quick coding tasks; no fixed task or message allowance is asserted here. |
| Go | $8/month | For lightweight coding tasks; availability can depend on account and region. |
| Plus | $20/month | The page explicitly lists web, CLI, IDE extension and iOS access plus cloud integrations. |
| Pro | $100, $200 or $500 USD/month | Choose the tier that fits your usage; the price does not establish a fixed number of completed tasks. |
| Business | $20/user/month with 2+ users when billed annually; $25/user/month when billed monthly | Workspace and administrative features; taxes, regional terms and total contract cost are not asserted. |
| Enterprise & Edu | Contact sales | No numeric price is published in this capture. |
| API key | Usage-based at standard API rates | Local CLI, SDK and IDE workflows; the pricing page says cloud features such as GitHub code review and Slack are unavailable on this route. |
What the subscription price covers
ChatGPT Work and Codex share pricing, credits and usage limits. Running more agents or starting Work tasks uses that shared allowance. A plan’s monthly price does not promise a fixed number of completed tasks: consumption varies with the model, context, reasoning, tools, retrieval and caching.
Additional credits can extend eligible plan usage. Their purchase price and applicable discounts depend on the plan or agreement. API-key usage follows separate OpenAI Platform billing at API rates; it does not use included ChatGPT plan credits. Check the account’s current balance, limits and billing terms before starting recurring or parallel work. This review did not buy credits, run a paid task or verify a billing charge.
The sign-in method changes what is available
The authentication guide separates ChatGPT subscription access from API-key usage-based access. The desktop app, CLI and IDE extension support both for local work, while Codex cloud requires ChatGPT sign-in. API-key usage is billed through the OpenAI Platform and follows API organization controls instead of ChatGPT workspace controls.
The CLI guide documents local repository inspection, edits, installed tools and codex exec automation. The IDE guide documents VS Code and compatible editors, plus separate Xcode and JetBrains integrations. The cloud guide documents isolated environments, repository selection and review before opening a pull request.
Treat permissions as a setup decision
OpenAI’s sandbox guide separates the technical sandbox from the approval policy. Local workspace-write can edit within the approved workspace and run routine commands; broader filesystem or network actions depend on the configured boundary and approval flow. Full access removes those sandbox restrictions. Review the active profile, repository scope, network access and credentials before assigning work.
Kingy has not purchased a plan, signed in, connected a repository, submitted code, run Codex, compared model output, measured task success, audited security or verified a billing charge for this update. The pricing, access and control statements above are first-party documentation claims, not hands-on results.
What remains unknown
- Only Pro has an explicitly stated USD currency in the retained pricing capture. No currency is inferred for the other quoted displays; region-wide availability, a tax-inclusive total and negotiated Enterprise/Edu prices remain unconfirmed.
- No fixed task, token or message allowance is asserted; OpenAI says usage depends on model, task size, local or cloud execution, context, reasoning, tools, retrieval and caching.
- Temporary referral offers and current usage estimates are omitted because they are not durable plan entitlements.
- No benchmark, code-quality score, security certification, market-share claim or recommendation is asserted.
Update sources and freshness
For the October 8 pricing recheck, Kingy compared retained HTTP 200 responses from the official pricing page and its Markdown response, captured at 05:36 UTC. Both explicitly list the three Pro tiers in USD. This body correction does not turn the other dollar displays into normalized currency amounts or establish a regional checkout total. The earlier September 20 pricing assessment remains historical evidence.
Kingy retained exact first-party Markdown responses and HTTP headers for pricing, authentication, CLI, IDE, cloud and sandboxing between 00:38:13 and 00:38:28 UTC on September 8, 2026. Any change to a bound price, plan name, access route or control statement requires a new source comparison and independent editorial review before publication.
Retained June 29 editorial note
Historical disclosure: The following June 29, 2026 text is retained unchanged as earlier editorial opinion. Its superlative comparisons and alternatives were not re-tested for this purchase-and-access update and should not be read as the current sourced comparison.
Codex is OpenAI’s software-engineering agent for working on coding tasks in isolated cloud environments and developer workflows.
Kingy AI Take
Codex matters because it moves OpenAI from coding assistance toward delegated software work. The strongest use case is not asking for a snippet; it is assigning bounded engineering tasks, reviewing the resulting diff, and using the agent to accelerate maintenance, tests, bug fixes, and implementation chores.
Best Use Cases
- Delegating small-to-medium code changes with clear acceptance criteria.
- Generating tests, migration notes, and implementation drafts.
- Investigating issues across a repository before a human reviews the fix.
- Comparing cloud-agent workflows against editor-first tools like Cursor.
Best Alternatives
Claude Code is the closest Anthropic alternative, Cursor is the strongest editor-native alternative, GitHub Copilot is the default enterprise IDE alternative, and Replit is better for browser-native app-building and hosting workflows.
What Feels Unproven
Codex still depends on clear task boundaries, repository permissions, review discipline, and the quality of test coverage around generated changes. Teams should verify current product availability, supported environments, model behavior, and security controls on OpenAI’s official Codex and developer pages before treating it as an autonomous engineering lane.
Related Kingy Coverage
- OpenAI’s Codex Boom Shows AI Agents Are No Longer Just for Developers
- Inside OpenAI Codex Remote GA and DigitalOcean Plugin
- How Kingy AI Explained OpenAI Codex
- Claude Code vs. Codex 2026
Official Sources
Last verified for this editorial note: June 29, 2026.
The Kingy Brief
Get The Kingy Brief.
AI changes, original tests and one practical thing to try. Fridays at 09:00 Vancouver time.
Free · Double opt-in · Unsubscribe anytime
Sign up for The Kingy Brief on Beehiiv.
Signup help and newsletter schedule
Signup form provided by Beehiiv. After submitting, check your inbox for "Confirm your subscription to The Kingy Brief" and open its confirmation link. Check Spam or Promotions if you cannot find it.
Fridays at 09:00 Vancouver time: source-checked AI changes, original tests and one practical thing to try. We skip a week when there is not enough verified material. Free. Unsubscribe anytime.
Tool Links
Featured in these videos
Launch History
Codex Cloud / reusable environments
Codex Cloud adds reusable development environments and cloud task execution across supported plans. Access: Cloud environments release.
Review Codex Cloud environment permissions and secrets. Repository access needs checking before reusing a shared cloud environment.
Codex Security Cloud
Codex Security Cloud enters research preview for repository investigation and proposed fixes in the cloud. Access: Research preview.
Codex Security Cloud requires repository permission review. Human review of proposed fixes remains necessary during research preview.
Codex CLI voice and agents view
Codex CLI gains voice task control and an agents view in a dated feature update. Access: CLI feature update.
Check Codex CLI microphone access and the installed version. Those checks determine whether voice control is usable in your development setup.
Codex Code Review desktop/cloud experience
Codex Code Review gains desktop and cloud experiences for reviewing project changes and proposed feedback. Access: Code review feature update.
Codex Code Review findings still need human review. Verify repository permissions before enabling automatic cloud reviews.
OpenAI releases Codex Security CLI and TypeScript SDK in limited beta
OpenAI published the Codex Security command-line client and TypeScript SDK. The CLI supports repository and change review, bulk scans, history, CI workflows, SARIF output and false-positive feedback. The…
The CLI and SDK make Codex Security easier to insert into repeatable engineering workflows and expose useful integration primitives such as typed findings, SARIF,…
OpenAI Codex Remote GA and DigitalOcean Plugin
OpenAI release notes list Codex Remote general availability and a DigitalOcean plugin update for Codex workflows.
- Launch readiness
- 6.9 / 10
- Demo evidence
- Not scored yet
- Creator-story fit
- Not scored yet
Score definitions and rubric
These are launch-record readiness heuristics, not product ratings.
Launch readiness
How complete and reviewable the launch record is, not the quality of the product.
Inputs and weights: Launch date 15%; qualifying source 10%; what launched 10%; demo 15%; category 10%; audience 10%; editorial assessment 10%; traction evidence 10%; creator or audience fit 10%.
Evidence inputs: Reviewed launch metadata, public source links, demo links, taxonomy, audience, editorial notes, and recorded traction signals.
Demo evidence
Whether the record contains useful, reviewable demonstration evidence; it is not a rating of product output quality.
Inputs and weights: Working demo URL 45%; video walkthrough 25%; clear description of what launched 10%; audience 10%; editorial assessment 10%.
Evidence inputs: Demo and video URLs plus the reviewed launch description, audience, and editorial notes.
Creator-story fit
Whether a launch has enough demonstrable evidence and audience relevance for a useful creator story; it does not predict views or guarantee coverage.
Inputs and weights: Demo evidence 25%; visual creator category 15%; audience 15%; editorial assessment 15%; traction evidence 10%; pricing clarity 10%; API or open-weight evidence 10%.
Evidence inputs: Reviewed demo, category, audience, editorial, traction, pricing, API, and open-weight fields.
- Scale
- 0.0–10.0. A present qualifying input receives its published weight; a missing input receives zero. Scores are rounded to one decimal.
- Assigned by
- Suggested by the deterministic field-completeness helper and assigned or approved by a Kingy editorial reviewer.
- Rubric and check date
- Rubric version P0-2026-08-10. The record’s “Last verified” date is the score check date. Checked: 2026-07-09.
- Confidence and missing data
- Confidence depends on source completeness. “Not scored yet” means no reviewed value; “Needs review” means the value or score set failed validation.
- Freshness
- Recalculate after a material launch, source, demo, pricing, audience, or traction change and during the record freshness review.
- Disputes
- Use “Suggest a correction” on the record and cite the relevant evidence. Commercial relationships cannot buy or alter a score.
OpenAI's release notes list Codex Remote reaching general availability alongside a DigitalOcean plugin update for Codex workflows (developers.openai.com). AI Engineers and Developers running coding-agent…
OpenAI Ona Acquisition for Codex
OpenAI announced an agreement to acquire Ona, subject to customary closing conditions, with the stated intent to bring Ona’s secure, persistent, customer-controlled cloud execution and orchestration technology into…
Ona gives OpenAI a credible route toward persistent, governed execution for Codex, but the announcement is a transaction and product-direction statement, not proof that…
OpenAI expands Codex for every role, tool, and workflow
OpenAI expanded Codex across roles, tools, and workflows, positioning it as a broader professional agent for software-adjacent and knowledge-work tasks.
- Launch readiness
- 6.1 / 10
- Demo evidence
- Not scored yet
- Creator-story fit
- Not scored yet
Score definitions and rubric
These are launch-record readiness heuristics, not product ratings.
Launch readiness
How complete and reviewable the launch record is, not the quality of the product.
Inputs and weights: Launch date 15%; qualifying source 10%; what launched 10%; demo 15%; category 10%; audience 10%; editorial assessment 10%; traction evidence 10%; creator or audience fit 10%.
Evidence inputs: Reviewed launch metadata, public source links, demo links, taxonomy, audience, editorial notes, and recorded traction signals.
Demo evidence
Whether the record contains useful, reviewable demonstration evidence; it is not a rating of product output quality.
Inputs and weights: Working demo URL 45%; video walkthrough 25%; clear description of what launched 10%; audience 10%; editorial assessment 10%.
Evidence inputs: Demo and video URLs plus the reviewed launch description, audience, and editorial notes.
Creator-story fit
Whether a launch has enough demonstrable evidence and audience relevance for a useful creator story; it does not predict views or guarantee coverage.
Inputs and weights: Demo evidence 25%; visual creator category 15%; audience 15%; editorial assessment 15%; traction evidence 10%; pricing clarity 10%; API or open-weight evidence 10%.
Evidence inputs: Reviewed demo, category, audience, editorial, traction, pricing, API, and open-weight fields.
- Scale
- 0.0–10.0. A present qualifying input receives its published weight; a missing input receives zero. Scores are rounded to one decimal.
- Assigned by
- Suggested by the deterministic field-completeness helper and assigned or approved by a Kingy editorial reviewer.
- Rubric and check date
- Rubric version P0-2026-08-10. The record’s “Last verified” date is the score check date. Checked: 2026-06-08.
- Confidence and missing data
- Confidence depends on source completeness. “Not scored yet” means no reviewed value; “Needs review” means the value or score set failed validation.
- Freshness
- Recalculate after a material launch, source, demo, pricing, audience, or traction change and during the record freshness review.
- Disputes
- Use “Suggest a correction” on the record and cite the relevant evidence. Commercial relationships cannot buy or alter a score.
OpenAI expanded Codex across roles, tools, and workflows, positioning it as a broader professional agent for software-adjacent knowledge work (openai.com). Developers, Product Managers, Data…
OpenAI releases the Codex app for managing multiple coding agents
OpenAI released the Codex app for macOS as a command center for running long-horizon and background coding-agent tasks, reviewing diffs, and using skills and automations.
The Codex app is important because it makes multi-agent software work feel manageable from a dedicated desktop surface.
Introducing Codex
OpenAI introduced Codex as a cloud-based software engineering agent that works on coding tasks in parallel in isolated environments.
A priority coding-agent record because Codex turned ChatGPT into an asynchronous code worker, not just an assistant.
Kingy coverage
Kingy AI articles about Codex, newest first.
- BlogCodex Composer Predictions for Pro Users: Access, Cost and Trade-Offs
- BlogHow to Connect Codex to Roblox Studio (MCP): Tested Setup
- BlogCodex Cloud Guide: Keep Building From Your Phone While Your Laptop Sleeps
- BlogCodex for Sales: The Living Field Guide to Agentic Revenue Work
- BlogCodex for Legal: The Living Field Guide to Agentic Legal Work
- BlogCodex Micro Review: Is OpenAI’s $230 Macropad Worth It?
- NewsCodex Record & Replay
- NewsCodex Reasoning Levels: Light, Medium, High, Extra High, Max & Ultra
More coverage (11)
- NewsOpenAI Just Brought Codex Computer Use to Windows: What PC Users Get Now
- NewsCodex Just Landed in the ChatGPT Mobile App: Inside OpenAI’s Push to Make AI Coding Truly Portable
- NewsOpenAI Codex Commands: CLI, App & IDE Cheat Sheet (2026)
- NewsOpenAI Daybreak Explained: Inside GPT‑5.5‑Cyber, Codex Security, and the New Frontier of AI Cyber Defense
- NewsOpenAI Codex Guide: Basics, Icons and Advanced Workflows
- NewsCodex /goal Guide: Persistent Objectives and Commands
- NewsCodex Gets a Memory: Inside OpenAI’s “Chronicle,” the Screen-Watching Agent Trying to End Context-Resetting
- NewsCodex vs Claude Code (2026): Models, Costs and Tested Tasks
- NewsOpenAI Codex Guide: ChatGPT desktop app, CLI, Skills and MCP
- NewsCodex by OpenAI: A Paradigm Shift in AI-Powered Coding
- NewsCodex CLI Demystified: What OpenAI’s Terminal-Based Coding Agent Actually Is and Why It Matters