OpenAI - Codex Docs Changes 2026-03-27 00:39 UTC → 2026-03-27 18:23 UTC

1# Upgrade your API integration | Codex use cases

2 

3[← All use cases](https://developers.openai.com/codex/use-cases)

4 

5Use Codex to update your existing OpenAI API integration to the latest recommended models and API features, while checking for regressions before you ship.

6 

7Intermediate

8 

91h

10 

11Related links

12 

13[Latest model guide](https://developers.openai.com/api/docs/guides/latest-model) [Prompt guidance](https://developers.openai.com/api/docs/guides/prompt-guidance) [OpenAI Docs MCP](/learn/docs-mcp) [Evals guide](https://developers.openai.com/api/docs/guides/evals)

14 

15## Best for

16 

17 - Teams upgrading from older models or API surfaces

18 - Repos that need behavior-preserving migrations with explicit validation

19 

20## Skills & Plugins

21 

22- [OpenAI Docs](https://github.com/openai/skills/tree/main/skills/.curated/openai-docs)

23 

24 Pull the current model, migration, and API guidance before Codex makes edits to your implementation.

25 

26## Starter prompt

27 

28Use $openai-docs to upgrade this OpenAI integration to the latest recommended model and API features.

29Specifically, look for the latest model and prompt guidance for this specific model.

30 Requirements:

31- Start by inventorying the current models, endpoints, and tool assumptions in the repo.

32- Identify the smallest migration plan that gets us onto the latest supported path.

33 - Preserve behavior unless a change is required by the new API or model.

34 - Update prompts using the latest model prompt guidance.

35- Call out any prompt, tool, or response-shape changes we need to review manually.

36 

37## Introduction

38 

39As we release new models and API features, we recommend upgrading your integration to benefit from the latest improvements.

40Changing from one model to another is often not as simple as just updating the model name.

41 

42There might be changes to the API–for example, for the GPT-5.4 model, we added a new `phase` parameter to the assistant message that is important to include in your integration–but most importantly, model behavior can be different and require changes to your existing prompts.

43 

44When migrating to a new model, you should make sure to not only make the necessary code changes, but also evaluate the impact on your workflows.

45 

46## Leverage the OpenAI Docs skill

47 

48All the specifics about the new API features and model behavior are documented in our docs, in the [latest model](https://developers.openai.com/api/docs/guides/latest-model) and [prompt guidance](https://developers.openai.com/api/docs/guides/prompt-guidance) guides.

49 

50The OpenAI Docs skill also includes [specific guidance](https://github.com/openai/codex/blob/6323f0104d17d211029faab149231ba787f7da37/codex-rs/skills/src/assets/samples/openai-docs/references/upgrading-to-gpt-5p4.md) as reference, codifying how to upgrade to the latest model–currently [GPT-5.4](https://developers.openai.com/api/docs/models/gpt-5.4).

51 

52Codex now automatically comes with the OpenAI Docs skill, so make sure to mention it in your prompt to access all the latest documentation and guidance when building with the OpenAI API.

53 

54## Build a robust evals pipeline

55 

56Codex can automatically update your prompts based on the latest prompt guidance, but you should have a way to automate verifying your integration is working as expected.

57 

58Make sure to build an evals pipeline that you can run every time you make changes to your integration, to verify there is no regression in behavior.

59 

60This [cookbook guide](https://developers.openai.com/cookbook/examples/evaluation/building_resilient_prompts_using_an_evaluation_flywheel) covers in detail how to do this using our [Evals API](https://developers.openai.com/api/docs/guides/evals).

61 

62## Related use cases

63 

64[![](/images/codex/codex-wallpaper-1.webp)

65 

66### Create browser-based games

67 

68Use Codex to turn a game brief into first a well-defined plan, and then a real browser-based...

69 

70Engineering Code](https://developers.openai.com/codex/use-cases/browser-games)[![](/images/codex/codex-wallpaper-1.webp)

71 

72### Bring your app to ChatGPT

73 

74Build one narrow ChatGPT app outcome end to end: define the tools, scaffold the MCP server...

75 

76Integrations Code](https://developers.openai.com/codex/use-cases/chatgpt-apps)[![](/images/codex/codex-wallpaper-3.webp)

77 

78### Build for iOS and macOS

79 

80Use Codex to scaffold SwiftUI projects, keep the build loop CLI-first with `xcodebuild` or...

81 

82Mobile Code](https://developers.openai.com/codex/use-cases/native-ios-macos-apps)

1# Create browser-based games | Codex use cases

2 

3Need

4 

5Backend stack

6 

7Default options

8 

9[Fastify](https://fastify.dev/) , WebSockets, [Postgres](https://www.postgresql.org/) , and [Redis](https://redis.io/)

10 

11Why it's needed

12 

13A strong default when the game needs persistence, matchmaking, leaderboards, or pub/sub.

1# Bring your app to ChatGPT | Codex use cases

2 

3Need

4 

5Widget framework

6 

7Default options

8 

9[React](https://react.dev/)

10 

11Why it's needed

12 

13A strong default for stateful widgets, especially when the UI needs filters, tables, or multi-step interaction.

1# Understand large codebases | Codex use cases

2 

3[← All use cases](https://developers.openai.com/codex/use-cases)

4 

5Use Codex to map unfamiliar codebases, explain different modules and data flow, and point you to the next files worth reading before you edit.

6 

7Easy

8 

95m

10 

11Related links

12 

13[Codex app](https://developers.openai.com/codex/app)

14 

15## Best for

16 

17 - New engineers onboarding to a new repo or service

18 - Anyone trying to understand how a feature works before changing it

19 

20## Starter prompt

21 

22Explain how the request flows through <name of the system area> in the codebase.

23 Include:

24 - which modules own what

25 - where data is validated

26 - the top gotchas to watch for before making changes

27 End with the files I should read next.

28 

29## Introduction

30 

31When you are new to a repo or dropped into an unfamiliar feature, Codex can help you get oriented before you start changing code. The goal is not just to get a high-level summary, but to map the request flow, understand which modules own what, and identify the next files worth reading.

32 

33## How to use

34 

35If you're new to a project, you can simply start by asking Codex to explain the whole codebase:

36 

37Explain this repo to me

38 

39If you need to contribute a new feature to an existing codebase, you can ask codex to explain a specific system area. The better you scope the request, the more concrete the explanation will be:

40 

411. Give Codex the relevant files, directories, or feature area you are trying to understand.

422. Ask it to trace the request flow and explain which modules own the business logic, transport, persistence, or UI.

433. Ask where validation, side effects, or state transitions happen before you edit anything.

444. End by asking which files you should read next and what the risky spots are.

45 

46A useful onboarding answer should leave you with a concrete map, not just a list of filenames. By the end, Codex should have explained the main flow, highlighted the risky parts, and pointed you to the next files or checks that matter before you start editing.

47 

48## Questions to ask next

49 

50Once Codex gives you a first pass, keep going until the explanation is specific enough that you would trust yourself to make the first edit. Good follow-up questions usually force it to call out assumptions, hidden dependencies, and the checks that matter after a change.

51 

52- Which module owns the actual business logic versus the transport or UI layer?

53- Where does validation happen, and what assumptions are enforced there?

54- What related files or background jobs are easy to miss if I change this flow?

55- Which tests or checks should I run after editing this area?

56 

57## Related use cases

58 

59[![](/images/codex/codex-wallpaper-3.webp)

60 

61### Iterate on difficult problems

62 

63Give Codex an evaluation system, such as scripts and reviewable artifacts, so it can keep...

64 

65Engineering Analysis](https://developers.openai.com/codex/use-cases/iterate-on-difficult-problems)[![](/images/codex/codex-wallpaper-1.webp)

66 

67### Create browser-based games

68 

69Use Codex to turn a game brief into first a well-defined plan, and then a real browser-based...

70 

71Engineering Code](https://developers.openai.com/codex/use-cases/browser-games)[![](/images/codex/codex-wallpaper-2.webp)

72 

73### Analyze datasets and ship reports

74 

75Use Codex to clean data, join sources, explore hypotheses, model results, and package the...

76 

77Data Analysis](https://developers.openai.com/codex/use-cases/datasets-and-reports)

1# Analyze datasets and ship reports | Codex use cases

2 

3Need

4 

5Analysis stack

6 

7Default options

8 

9[pandas](https://pandas.pydata.org/) with [matplotlib](https://matplotlib.org/) or [seaborn](https://seaborn.pydata.org/)

10 

11Why it's needed

12 

13Good defaults for import, profiling, joins, cleaning, and the first round of charts.

1# Turn Figma designs into code | Codex use cases

2 

3Need

4 

5Design source

6 

7Default options

8 

9[Figma](https://www.figma.com/)

10 

11Why it's needed

12 

13A concrete frame or component selection keeps the implementation grounded.

1# Build responsive front-end designs | Codex use cases

2 

3[← All use cases](https://developers.openai.com/codex/use-cases)

4 

5Use Codex to translate screenshots and design briefs into code that matches the repo's design system, then use Playwright to compare the implementation to your references for different screen sizes and iterate until it looks right.

6 

7Intermediate

8 

91h

10 

11Related links

12 

13[Codex skills](https://developers.openai.com/codex/skills)

14 

15## Best for

16 

17 - Creating new front-end projects from scratch

18- Implementing already designed screens or flows from screenshots in an existing codebase

19 

20## Skills & Plugins

21 

22- [Playwright](https://github.com/openai/skills/tree/main/skills/.curated/playwright-interactive)

23 

24 Open the app in a real browser to verify the implementation and iterate on layout and behavior.

25 

26## Starter prompt

27 

28Implement this UI in the current project using the screenshots and notes I provide as the source of truth.

29 Requirements:

30 - Reuse the existing design system components and tokens.

31- Translate the screenshots into this repo's utilities and component patterns instead of inventing a parallel system.

32 - Match spacing, layout, hierarchy, and responsive behavior closely.

33 - Respect the repo's routing, state, and data-fetch patterns.

34 - Make the page responsive on desktop and mobile.

35- If any screenshot detail is ambiguous, choose the simplest implementation that still matches the overall direction and note the assumption briefly.

36 Validation:

37- Compare the finished UI against the provided screenshots for both look and behavior.

38- Use $playwright-interactive to check that the UI matches the references and iterate as needed until it does.

39 

40## Introduction

41 

42When you have screenshots, a short design brief, or a few references for inspiration, Codex can turn those into responsive UI without ignoring the patterns already established in your project.

43 

44With the Playwright skill, Codex can open the app in a real browser, compare the implementation to your screenshots for different screen sizes, and iterate on layout or behavior until the result is closer to the target.

45 

46## Start from references

47 

48Give Codex the clearest references you have for the UI you want. A single screenshot can be enough for a narrow task, but the handoff gets better when you include multiple states such as desktop and mobile layouts, hover or selected states, and any empty or loading views that matter.

49 

50The references do not need to be perfect design deliverables. They just need to make the intended hierarchy, spacing, and direction concrete enough that Codex is not guessing.

51 

52## Be specific

53 

54The more specific you are about the expected interaction patterns and the style you want, the better the result will be.

55The model tends to default to high-frequency patterns and style so if it's not obvious from your references that you want something else, the UI might look generic.

56The more input you give, be it more reference inspiration or more specific instructions, the more you can expect to have a UI that stands out.

57 

58## Prepare the design system

59 

60Codex works best when the target repo already has a clear component layer. Codex can automatically use your existing component and design system instead of recreating them from scratch.

61 

62If you think it's necessary (i.e. if you're not using a standard stack), specify to Codex which primitives to reuse, where your tokens live, and what the repo considers canonical for buttons, inputs, cards, typography, and icons.

63 

64If you're starting from an existing codebase, it's very likely that Codex will understand on its own how to use your components and design system, but if starting from scratch, it's a good idea to be explicit.

65 

66Ask Codex to treat the screenshots as a visual target but to translate that target into the project's actual utilities, component wrappers, color system, typography scale, spacing tokens, routing, state management, and data-fetch patterns.

67 

68## Leverage Playwright

69 

70Playwright is a great tool to help Codex iterate on the UI. With it, Codex can open the app in a real browser, compare the implementation to the screenshots you provided, and iterate on layout or behavior.

71 

72It can resize the browser window to different screen sizes and check the layout at different breakpoints.

73 

74Make sure you have the Playwright interactive skill enabled in Codex. For more details, see the [skills documentation](https://developers.openai.com/codex/skills).

75 

76## Iterate

77 

78The first pass should already be directionally close to the screenshots. For complex layouts, interactions, or animation-heavy UI, expect a few rounds of adjustment.

79 

80Ask Codex to compare the implementation back to the screenshots, not just whether the page builds. When conflicts come up, it should prefer the repo's design-system tokens and only make minimal spacing or sizing adjustments needed to preserve the overall look of the design.

81 

82Use additional screenshots or short notes if they help clarify states that are not obvious from one image.

83 

84### Suggested follow-up prompt

85 

86[current implementation image] [reference image]

87This doesn't look right. Make sure to implement something that matches closely the reference:

88[if needed, specify what is different]

89 

90## Related use cases

91 

92[![](/images/codex/codex-wallpaper-2.webp)

93 

94### Turn Figma designs into code

95 

96Use Codex to pull design context, assets, and variants from Figma, translate them into code...

97 

98Front-end Design](https://developers.openai.com/codex/use-cases/figma-designs-to-code)[![](/images/codex/codex-wallpaper-3.webp)

99 

100### Generate slide decks

101 

102Use Codex to update existing presentations or build new decks by editing slides directly...

103 

104Data Automation](https://developers.openai.com/codex/use-cases/generate-slide-decks)[![](/images/codex/codex-wallpaper-1.webp)

105 

106### Create browser-based games

107 

108Use Codex to turn a game brief into first a well-defined plan, and then a real browser-based...

109 

110Engineering Code](https://developers.openai.com/codex/use-cases/browser-games)

1# Generate slide decks | Codex use cases

2 

3[← All use cases](https://developers.openai.com/codex/use-cases)

4 

5Use Codex to update existing presentations or build new decks by editing slides directly through code, generating visuals, and applying repeatable layout rules slide by slide.

6 

7Easy

8 

930m

10 

11Related links

12 

13[Image generation guide](https://developers.openai.com/api/docs/guides/image-generation)

14 

15## Best for

16 

17 - Teams turning notes or structured inputs into repeatable slide decks

18 - Creating new visual presentations from scratch

19- Rebuilding or extending decks from screenshots, PDFs, or reference presentations

20 

21## Skills & Plugins

22 

23- [Slides](https://github.com/openai/skills/tree/main/skills/.curated/slides)

24 

25 Create and edit `.pptx` decks in JavaScript with PptxGenJS, bundled helpers, and render and validation scripts for overflow, overlap, and font checks.

26- [ImageGen](https://github.com/openai/skills/tree/main/skills/.curated/imagegen)

27 

28 Generate illustrations, cover art, diagrams, and slide visuals that match one reusable visual direction.

29 

30## Starter prompt

31 

32Use $slides with $imagegen to edit this slide deck in the following way:

33 - If present, add logo.png in the bottom right corner on every slide

34- On slides X, Y and Z, move the text to the left and use image generation to generate an illustration (style: abstract, digital art) on the right

35- Preserve text as text and simple charts as native PowerPoint charts where practical.

36 - Add these slides: [describe new slides here]

37- Use the existing branding on new slides and new text (colors, fonts, layout, etc.)

38- Render the updated deck to slide images, review the output, and fix layout issues before delivery.

39- Run overflow and font-substitution checks before delivery, especially if the deck is dense.

40- Save reusable prompts or generation notes when you create a batch of related images.

41 Output:

42 - A copy of the slide deck with the changes applied

43 - notes on which slides were generated, rewritten, or left unchanged

44 

45## Introduction

46 

47You can use Codex to manipulate PowerPoint decks in a systematic way, using the Slides skill to create and edit decks with PptxGenJS, and using image generation to generate visuals for the slides.

48 

49Skills can be installed directly from the Codex app–see our [skills documentation](https://developers.openai.com/codex/skills) for more details.

50 

51You can create new decks from scratch, describing what you want, but the ideal workflow is to start from an existing deck–already set up with your branding guidelines–and ask Codex to edit it.

52 

53## Start from the source deck and references

54 

55If a deck already exists, ask Codex to inspect it before making changes.

56 

57The slides skill is opinionated here: match the source aspect ratio before you rebuild layout, and default to 16:9 only when the source material does not already define the deck size. If the references are screenshots or a PDF, ask Codex to render or inspect them first so it can compare slide geometry visually instead of guessing.

58 

59## Keep the deck editable

60 

61When building out new slides, ask Codex to keep the slides editable: when slides contain text, charts, or simple layout elements, those should stay PowerPoint-native when practical. Text should stay text. Simple bar, line, pie, and histogram visuals should stay native charts when possible. For diagrams or visuals that are too custom for native slide objects, Codex can generate or place SVG and image assets deliberately instead of rasterizing the whole slide.

62 

63For example, if you want to build a complex timeline with illustrations, instead of generating a whole image, ask Codex to generate each illustration separately (using a set style prompt as reference), place them on the slide, then link them using native lines. The text and dates should be text objects as well, and not included in the illustrations.

64 

65## Generate visuals intentionally

66 

67Image generation is most useful when the slides need a cover image, a concept illustration, or a lightweight diagram that would otherwise take manual design work. Ask Codex to define the visual direction first, then reuse that direction consistently across the whole deck.

68 

69When several slides need related visuals, have Codex save the prompts or generation notes it used. That makes the deck easier to extend later without starting over stylistically.

70 

71## Keep slide logic explicit

72 

73Deck automation works better when Codex treats each slide as its own decision. Some slides should preserve exact copy, some need a stronger headline and cleaner structure, and some should stay mostly untouched apart from asset cleanup or formatting fixes.

74 

75The slides skill also ships with bundled layout helpers. Ask Codex to copy those helpers into the working directory and reuse them instead of reimplementing spacing, text-sizing, and image-placement logic on every deck.

76 

77## Validation before delivery

78 

79Decks are easy to get almost right and still ship with clipped text, substituted fonts, or layout drift that only shows up after export. The slides skill includes scripts to render decks to per-slide PNGs, build a quick montage for review, detect overflow beyond the slide canvas, and report missing or substituted fonts.

80 

81Ask Codex to use those checks before it hands back the final deck, especially when slides are dense or margins are tight.

82 

83## Example ideas

84 

85Here are some ideas you could try with this use case:

86 

87### New deck from scratch

88 

89You can create new slide decks from scratch, describing what you want slide by slide and the overall vibe.

90If you have assets like logos or images, you can copy them in the same folder so that Codex can easily access them.

91 

92Create a new slide deck with the following slides:

93- Slide 1: Title slide with the company logo (logo.png) and the title of the presentation

94- Slide 2: Agenda slide with the key points of the presentation

95- Slide 3: [TITLE] [TAGLINE] [DESCRIPTION]

96- ...

97- Slide N: Conclusion slide with the key takeaways

98- Slide N+1: Q&A slide with my picture (my-picture.png)

99 

100### Deck template update

101 

102You can update a deck template on a regular basis (weekly, monthly, quarterly, etc.) with new content.

103If you're doing this frequently, create a file like `guidelines.md` to define the content and structure of the deck and how it should be updated.

104 

105Combine it with other skills to fetch information from your preferred data

106 sources.

107 

108For example, if you need to give quarterly updates to your stakeholders, you can update the deck template with new numbers and insights.

109 

110Update the deck template, pulling content from [integration 1] and [integration 2].

111Make sure to follow guidelines defined in guidelines.md.

112 

113### Adjust existing deck

114 

115If you built a deck but want to adjust it to fix spacing, misaligned text, or other layout issues, you can ask Codex to fix it.

116 

117Adjust the deck to make sure the following layout rules are followed:

118- Spacing should be consistent when there are multiple items on the same slide displayed in a row or grid.

119- When there are multiple items on the same slide displayed in a row or grid, the items are aligned horizontally or vertically depending on the content.

120- All text boxes should be aligned left, except when they are below an illustration

121- All titles should use the font [font name] and size [size]

122- All captions should be in [color]

123- ....

124 

125## Related use cases

126 

127[![](/images/codex/codex-wallpaper-2.webp)

128 

129### Kick off coding tasks from Slack

130 

131Mention `@Codex` in Slack to start a task tied to the right repo and environment, then...

132 

133Integrations Automation](https://developers.openai.com/codex/use-cases/slack-coding-tasks)[![](/images/codex/codex-wallpaper-2.webp)

134 

135### Analyze datasets and ship reports

136 

137Use Codex to clean data, join sources, explore hypotheses, model results, and package the...

138 

139Data Analysis](https://developers.openai.com/codex/use-cases/datasets-and-reports)[![](/images/codex/codex-wallpaper-2.webp)

140 

141### Build responsive front-end designs

142 

143Use Codex to translate screenshots and design briefs into code that matches the repo's...

144 

145Front-end Design](https://developers.openai.com/codex/use-cases/frontend-designs)

1# Review pull requests faster | Codex use cases

2 

3[← All use cases](https://developers.openai.com/codex/use-cases)

4 

5Use Codex in GitHub to automatically surface regressions, missing tests, and documentation issues directly on a pull request.

6 

7Easy

8 

95s

10 

11Related links

12 

13[Use Codex in GitHub](https://developers.openai.com/codex/integrations/github) [Custom instructions with AGENTS.md](https://developers.openai.com/codex/guides/agents-md)

14 

15## Best for

16 

17 - Teams that want another review signal before human merge approval

18 - Large codebases for projects in production

19 

20## Skills & Plugins

21 

22- [Security Best Practices](https://github.com/openai/skills/tree/main/skills/.curated/security-best-practices)

23 

24 Focus the review on risky surfaces such as secrets, auth, and dependency changes.

25 

26## Starter prompt

27 

28@codex review for security regressions, missing tests, and risky behavior changes.

29 

30## How to use

31 

32Start by adding Codex code review to your GitHub organization or repository. See [Use Codex in GitHub](https://developers.openai.com/codex/integrations/github) for more details.

33 

34You can set up Codex to automatically review every pull request, or you can request a review with `@codex review` in a pull request comment.

35 

36If Codex flags a regression or potential issue, you can ask it to fix it by commenting on the pull request with a follow-up prompt like `@codex fix it`.

37 

38This will start a new cloud task that will fix the issue and update the pull request.

39 

40## Define additional guidance

41 

42To customize what Codex reviews, add or update a top-level `AGENTS.md` with a section like this:

43 

44```md

45## Review guidelines

46 

47- Flag typos and grammar issues as P0 issues.

48- Flag potential missing documentation as P1 issues.

49- Flag missing tests as P1 issues.

50 ...

51```

52 

53Codex applies guidance from the closest `AGENTS.md` to each changed file. You can place more specific instructions deeper in the tree when particular packages need extra scrutiny.

54 

55## Related use cases

56 

57[![](/images/codex/codex-wallpaper-1.webp)

58 

59### Bring your app to ChatGPT

60 

61Build one narrow ChatGPT app outcome end to end: define the tools, scaffold the MCP server...

62 

63Integrations Code](https://developers.openai.com/codex/use-cases/chatgpt-apps)[![](/images/codex/codex-wallpaper-3.webp)

64 

65### Generate slide decks

66 

67Use Codex to update existing presentations or build new decks by editing slides directly...

68 

69Data Automation](https://developers.openai.com/codex/use-cases/generate-slide-decks)[![](/images/codex/codex-wallpaper-2.webp)

70 

71### Kick off coding tasks from Slack

72 

73Mention `@Codex` in Slack to start a task tied to the right repo and environment, then...

74 

75Integrations Automation](https://developers.openai.com/codex/use-cases/slack-coding-tasks)

1# Iterate on difficult problems | Codex use cases

2 

3[← All use cases](https://developers.openai.com/codex/use-cases)

4 

5Give Codex an evaluation system, such as scripts and reviewable artifacts, so it can keep improving a hard task until the scores are good enough.

6 

7Advanced

8 

9Long-running

10 

11Related links

12 

13[Custom instructions with AGENTS.md](https://developers.openai.com/codex/guides/agents-md) [Codex workflows](https://developers.openai.com/codex/workflows)

14 

15## Best for

16 

17- Problems where each iteration can be scored, but the best result usually takes many passes

18- Tasks with visual or subjective outputs that need both deterministic checks and an LLM-as-a-judge score

19- Long-running Codex sessions where you want progress tracked clearly instead of relying on context

20 

21## Starter prompt

22 

23I have a difficult task in this workspace and I want you to run it as an eval-driven improvement loop.

24 Before changing anything:

25 - Read `AGENTS.md`.

26 - Find the script or command that scores the current output.

27 Iteration loop:

28 - Make one focused improvement at a time.

29 - Re-run the eval command after each meaningful change.

30 - Log the scores and what changed.

31- Inspect generated artifacts directly. If the output is visual, use `view\_image`.

32 - Keep going until both the overall score and the LLM average are above 90%.

33 Constraints:

34 - Do not stop at the first acceptable result.

35- Do not revert to an earlier version unless the new result is clearly worse in scores or artifacts.

36- If the eval improves but is still below target, explain the bottleneck and continue.

37 Output:

38 - current best scores

39 - log of major iterations

40 - remaining risks or weak spots

41 

42## Introduction

43 

44Some tasks are easy to verify in one shot: the build passes, the tests go green, and you are done. But there are some optimization problems that are difficult to solve, and need many iterations with a tight evaluation loop. To know which direction to go in, Codex needs to inspect the current output, score it, decide the next change, and repeat until the result is actually good.

45 

46This type of use case pairs well with a custom UI that lets you inspect progress visually, by having Codex log the outputs and generated artifacts for each iteration.

47You can watch Codex continue working in the app while the target artifact, model output, or generated asset keeps improving.

48The key is to give Codex the necessary scripts to generate the evaluation metrics and the artifacts to inspect.

49 

50## Start with evals

51 

52Before the task begins, define how success will be measured. The best setup usually combines:

53 

54- **Deterministic checks:** things the scripts can score directly, such as constraint violations or deterministic metrics computed with code

55- **LLM-as-a-judge checks:** rubric-based scores for qualities that are harder to encode exactly, such as resemblance, readability, usefulness, or overall quality - this can rely on text or image outputs

56 

57If the subjective part matters, give Codex a script that can call a model for example using the [Responses API](https://developers.openai.com/api/reference/resources/responses/methods/create) and return structured scores. The point is not to replace deterministic checks, it's to supplement them with a consistent judge for the part humans would otherwise assess by eye.

58 

59The loop works best when the eval output is machine-readable, saved after every run, and easy to compare over time.

60 

61**Tip**: Ask Codex to generate the evaluation script for you, describing the

62 checks you want to run.

63 

64## Give Codex a stopping rule

65 

66Hard tasks often drift because the prompt says “keep improving” without saying when to stop. Make the stopping rule explicit.

67 

68A practical pattern is:

69 

701. Set a target for the overall score.

712. Set a separate target for the LLM-judge average.

723. Tell Codex to continue until both are above the threshold, not just one.

73 

74For example, if the goal is a high-quality artifact, ask Codex to keep going until both the overall score and the LLM average are above 90%. That makes the task legible: Codex can tell whether it is still below target, where the gap is, and whether the latest change helped.

75 

76## Keep a running log of the loop

77 

78Long-running work is much more reliable when Codex keeps notes about the loop instead of trying to remember everything from the thread.

79 

80That running log should record:

81 

82- the current best scores

83- what changed on the last iteration

84- what the eval said got better or worse

85- what Codex plans to try next

86 

87This is especially important when the task runs for a long time. The log becomes the handoff point for the next session and the self-evaluation record for the current one.

88 

89## Inspect the artifact, not just the logs

90 

91For some difficult tasks, the code diff and metric output are not enough. Codex should look at the artifact it produced.

92 

93If the output is visual, such as a generated image, layout, or rendered state, let Codex inspect that artifact directly, for example when the output lives on disk as an image and compare the current result to the prior best result or to the intended rubric.

94 

95This makes the loop stronger:

96 

97- the eval script reports the score

98- the artifact shows what the score missed

99- the next change is grounded in both

100 

101That combination is much more effective than changing code blindly between runs.

102 

103## Make every iteration explicit

104 

105Ask Codex to follow the same loop every time:

106 

1071. Run the evals on the current baseline.

1082. Identify the biggest failure mode from the scores and artifacts.

1093. Make one focused change that addresses that bottleneck.

1104. Re-run the evals.

1115. Log the new scores and whether the change helped.

1126. Continue until the thresholds are met.

113 

114This discipline matters. If each iteration changes too many things at once, Codex cannot tell which idea improved the score. If it skips logging, the session becomes hard to trust and hard to resume.

115 

116## Related use cases

117 

118[![](/images/codex/codex-wallpaper-1.webp)

119 

120### Understand large codebases

121 

122Use Codex to map unfamiliar codebases, explain different modules and data flow, and point...

123 

124Engineering Analysis](https://developers.openai.com/codex/use-cases/codebase-onboarding)[![](/images/codex/codex-wallpaper-1.webp)

125 

126### Create browser-based games

127 

128Use Codex to turn a game brief into first a well-defined plan, and then a real browser-based...

129 

130Engineering Code](https://developers.openai.com/codex/use-cases/browser-games)[![](/images/codex/codex-wallpaper-2.webp)

131 

132### Analyze datasets and ship reports

133 

134Use Codex to clean data, join sources, explore hypotheses, model results, and package the...

135 

136Data Analysis](https://developers.openai.com/codex/use-cases/datasets-and-reports)

1# Build for iOS and macOS | Codex use cases

2 

3Need

4 

5Project automation

6 

7Default options

8 

9[XcodeBuildMCP](https://www.xcodebuildmcp.com/)

10 

11Why it's needed

12 

13A strong option once you need Codex to inspect schemes and targets, launch the app, capture screenshots, and keep iterating without leaving the agentic loop.

1# Kick off coding tasks from Slack | Codex use cases

2 

3[← All use cases](https://developers.openai.com/codex/use-cases)

4 

5Mention `@Codex` in Slack to start a task tied to the right repo and environment, then review the result back in the thread or in Codex cloud.

6 

7Easy

8 

95m

10 

11Related links

12 

13[Use Codex in Slack](https://developers.openai.com/codex/integrations/slack) [Codex cloud environments](https://developers.openai.com/codex/cloud/environments)

14 

15## Best for

16 

17- Async handoffs that start in a Slack thread and already have enough context to act on

18- Teams that want quick issue triage, bug fixes, or scoped implementation work without context switching

19 

20## Starter prompt

21 

22@Codex analyze the issue mentioned in this thread and implement a fix in <name of your environment>.

23 

24## How to use

25 

261. Install the Slack app, connect the right repositories and environments, and add `@Codex` to the channel.

272. Mention `@Codex` in a thread with a clear request, constraints, and the outcome you want.

283. Open the task link, review the result, and continue the follow-up in Slack if the task needs another pass.

29 

30You can learn more about how to use Codex in Slack in the [dedicated guide](https://developers.openai.com/codex/integrations/slack).

31 

32## Tips

33 

34- If the thread does not already include enough context or suggested fix, include in your prompt some guidance

35- Make sure the repo and environment mapping are correct by mentioning the name of the project or environment in your prompt

36- Scope the request so Codex can finish it without a second planning loop

37- If your project is a large codebase, guide Codex by mentioning which files or folders are relevant to the task

38 

39## Related use cases

40 

41[![](/images/codex/codex-wallpaper-3.webp)

42 

43### Generate slide decks

44 

45Use Codex to update existing presentations or build new decks by editing slides directly...

46 

47Data Automation](https://developers.openai.com/codex/use-cases/generate-slide-decks)[![](/images/codex/codex-wallpaper-2.webp)

48 

49### Analyze datasets and ship reports

50 

51Use Codex to clean data, join sources, explore hypotheses, model results, and package the...

52 

53Data Analysis](https://developers.openai.com/codex/use-cases/datasets-and-reports)[![](/images/codex/codex-wallpaper-1.webp)

54 

55### Bring your app to ChatGPT

56 

57Build one narrow ChatGPT app outcome end to end: define the tools, scaffold the MCP server...

58 

59Integrations Code](https://developers.openai.com/codex/use-cases/chatgpt-apps)