main.js · part 225

60 text occurrences from desktop/Cursor.app/Contents/Resources/app/extensions/cursor-agent-exec/dist/main.js, part 225. Every entry preserves the shipped literal and its saved verdict or selection reason.

File contents and all parts · All files

Unable to generate an image from this description. Try rewording it.

The image provider is temporarily unavailable (status ${t}). Try again in a moment.

${void 0!==t?t:"Image generation request rejected by provider"}${void 0!==r?` (code=${r})`:""}

Image generation request rejected by provider

 (code=${r})

status=${r} ${n}

${t}. ${r}

[generate-image] ref image read failed

[generate-image] ref image read success

[generate-image] ref image read empty

[generate-image] ref image read failed

Failed to save generated image: ${r}

permission denied (operation=${p.result.value.operation}, isReadonly=${p.result.value.isReadonly}, error=${p.result.value.error})

Failed to save generated image: ${r}

rejected (reason=${e})

Failed to save generated image: ${r}

Failed to find tool call context

No space left on device

Failed to save generated image: ${r}

Failed to save generated image: ${r}

Failed to save generated image: ${r}

unknown result case

Failed to save generated image: ${r}

unhandled case: ${String(e.case)}

Failed to save generated image: ${r}

[generate-image] write failed

[generate-image] write success

Unhandled version: ${e}

Generate an image file from a text description.

STRICT INVOCATION RULES (must follow):
- Only use this tool when the user explicitly asks for an image. Do not generate images "just to be helpful".
- Do not use this tool for data heavy visualizations such as charts, plots, tables.

General guidelines:
- Provide a concrete description first: subject(s), layout, style, colors, text (if any), and constraints.
- If the user requests an aspect ratio, set `aspect_ratio` to one of "1:1", "4:3", "3:4", "16:9", or "9:16".
- If the user provides reference images, include them in `reference_image_paths`.
- Do not repeat generated images as Markdown in your response; the client displays tool-generated images automatically.

Examples that should call this tool:
- user: "Generate an app icon for a note-taking app, minimal flat vector style." (explicitly requests an image asset)
- user: "Make a UI mockup of a settings screen with a dark mode toggle." (explicitly requests a UI mockup)
- user: "Generate an asset of a game character with a sword." (explicitly requests a visual asset)

Examples that should not call this tool:
- user: "Create a plan to refactor this module." (planning request; respond in text or mermaid diagram)
- user: "Generate a chart of sales and revenue using data.csv." (data visualization; generate via code)

Unhandled version: ${e}

GenerateImage is not available for the current selected model. If important to generate an image (e.g. the user asked for it), then ask the user to switch models.

[generate-image] short/suspicious prompt rejected

Image generation was not performed — the tool call appears to be unintended. Do not retry or call GenerateImage again unless the user explicitly asks for an image.

[generate-image] no project folder available to save the generated image

generate_image needs a workspace/project folder so it has somewhere to save the PNG. Open a folder or run from a workspace, then try again.

[generate-image] aborted

Unhandled finish stage: ${e}

Successfully generated image. Display it in your response using:
<img src="${e}" alt="Generated image" />

Successfully generated image at: ${e}
Always use this absolute path when referring to the image. Do not repeat this image as a Markdown reference; it is already displayed to the user.

Failed to generate image: no image data returned

Successfully generated image at: ${e}
Always use this absolute path when referring to the image. Do not repeat this image as a Markdown reference; it is already displayed to the user.

Successfully generated image at: ${e.result.value.filePath}
Always use this absolute path when referring to the image. Do not repeat this image as a Markdown reference; it is already displayed to the user.

Failed to generate image, error: ${e.result.value.error}

An unknown error occurred.

<!-- cursor:tour-artifact

${e.slice(0,t).join(",")} +${e.length-t} more

PR code tour reads are unavailable in this session.

Object expected.

Symbol.asyncDispose is not defined.

Symbol.dispose is not defined.

Object not disposable.

An error was suppressed during disposal.

Natural-language instructions for regenerating the PR code tour, such as 'make it shorter' or 'focus on review risks'. Do not combine with markdown or artifactPath.

COMPLETE revision id to edit in place. Required for replace, section patch, and attach. From <pr_code_tour_context> or GetPrCodeTour.

Replacement markdown for the whole tour, or the new section body when heading is also set.

ATX heading text or slug of the target section. With markdown: replace that section only. With artifactPath: attach the media into that file-anchored section (defaults to the last file-anchored section).

Media artifact to attach to the revision as visible, playable step media: an image (png/jpeg/gif/webp) or a recording (.mp4/.webm), given as a workspace artifact path (e.g. /opt/cursor/artifacts/demo.mp4) or an artifact URL this agent owns. Images must already be uploaded (walkthrough artifacts upload when referenced in a response or PR body). Recordings upload from the VM on attach, so a recording saved under the artifacts directory can be attached right after it is captured. Provide with revisionId. The attach fails with an actionable error rather than storing media a reader cannot see.

Optional alt text for an attached artifact; rendered as the media's caption.

Optional commit-subset base SHA. Provide together with headSha to generate the tour from the baseSha..headSha range instead of the full PR diff.

Optional commit-subset head SHA. Provide together with baseSha to generate the tour from the baseSha..headSha range instead of the full PR diff.