@mindstudio-ai/agent 0.1.182 → 0.1.183
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- package/CHANGELOG.md +7 -0
- package/dist/cli.js +11 -9
- package/dist/generated/steps.d.ts +1 -1
- package/dist/index.js +1 -1
- package/dist/index.js.map +2 -2
- package/llms.txt +1 -1
- package/package.json +1 -1
package/CHANGELOG.md
CHANGED
|
@@ -4,6 +4,13 @@ What changed in `@mindstudio-ai/agent`, newest first, each version under its own
|
|
|
4
4
|
|
|
5
5
|
Versions before 0.1.163 aren't recorded here.
|
|
6
6
|
|
|
7
|
+
## 0.1.183
|
|
8
|
+
|
|
9
|
+
### Changed behavior
|
|
10
|
+
|
|
11
|
+
- `mindstudio login` waits for the browser's approval for as long as the platform keeps the sign-in open, which the platform now says when the sign-in starts: three minutes. It waited two minutes whatever the platform's own limit was, and against a platform that doesn't say it still does. Check: nothing in app code. A script that drives `mindstudio login` and expects it to give up after two minutes should allow three.
|
|
12
|
+
- `extractText`'s usage notes say which size limit applies to a file: 50 MB for a URL that isn't one of the app's own files, and for a kind read without a model (text, HTML, spreadsheets, office documents); the model's own limit for one of the app's files that a model reads. They used to say both "up to 100 MB" for LlamaParse and "50MB per URL". Nothing changed at run time. Check: nothing.
|
|
13
|
+
|
|
7
14
|
## 0.1.182
|
|
8
15
|
|
|
9
16
|
### New
|
package/dist/cli.js
CHANGED
|
@@ -320,7 +320,7 @@ var init_metadata = __esm({
|
|
|
320
320
|
"extractText": {
|
|
321
321
|
stepType: "extractText",
|
|
322
322
|
description: "Download a file from a URL and extract its text content. Supports PDFs (including scanned/image-based PDFs via OCR), plain text files, and other document formats.",
|
|
323
|
-
usageNotes: "- Best suited for PDFs and raw text/document files. For web pages, use the scrapeUrl step instead.\n- Handles both text-layer PDFs and image-based/scanned PDFs (e.g. Figma/Canva exports, scanned documents). Image-based PDFs are processed with OCR automatically \u2014 there is no need to convert PDF pages to images first.\n- Plain text, code, HTML, spreadsheets and Office documents (docx, pptx, odt, rtf, epub) are read without a model and cost nothing. PDFs and images go to the extraction model, billed per page.\n- Accepts a single URL, a comma-separated list of URLs, or a JSON array of URLs.\n- Nothing is published. One of the app's own files \u2014 including a private one, or a share link \u2014 is read where it sits; any other URL is downloaded and staged privately for the extraction model, then discarded when extraction finishes.\n- `text` is each document's text: a string for one URL, an array for several. `documents` has each document page by page, with its page numbers and the pages billed. Set `includeBlocks` to add each page's typed blocks (title, text, table, image, header, footer\u2026) with their bounding boxes, and the page's size.\n- A file that can't be read (unreachable, an unsupported type, past the model's limits, or refused by the model) fails the step with the reason.\n- `extractionModelOverride.model` picks the model; omit it to use the workspace's default:\n - `mistral-ocr-4.1` (the default): PDFs and images up to 50 MB and 1,000 pages, with typed blocks.\n - `mistral-document-ai-2512`: Mistral Document AI on Microsoft Foundry, up to 30 MB and 30 pages, with typed blocks.\n - `llamaparse`: LlamaParse's agentic mode, up to 100 MB and 1,000 pages, markdown without blocks.\n - `google-document-ai`: Google Document AI OCR, up to 500 pages, where the deployment has set it up.\n-
|
|
323
|
+
usageNotes: "- Best suited for PDFs and raw text/document files. For web pages, use the scrapeUrl step instead.\n- Handles both text-layer PDFs and image-based/scanned PDFs (e.g. Figma/Canva exports, scanned documents). Image-based PDFs are processed with OCR automatically \u2014 there is no need to convert PDF pages to images first.\n- Plain text, code, HTML, spreadsheets and Office documents (docx, pptx, odt, rtf, epub) are read without a model and cost nothing. PDFs and images go to the extraction model, billed per page.\n- Accepts a single URL, a comma-separated list of URLs, or a JSON array of URLs.\n- Nothing is published. One of the app's own files \u2014 including a private one, or a share link \u2014 is read where it sits; any other URL is downloaded and staged privately for the extraction model, then discarded when extraction finishes.\n- `text` is each document's text: a string for one URL, an array for several. `documents` has each document page by page, with its page numbers and the pages billed. Set `includeBlocks` to add each page's typed blocks (title, text, table, image, header, footer\u2026) with their bounding boxes, and the page's size.\n- A file that can't be read (unreachable, an unsupported type, past the model's limits, or refused by the model) fails the step with the reason.\n- `extractionModelOverride.model` picks the model; omit it to use the workspace's default:\n - `mistral-ocr-4.1` (the default): PDFs and images up to 50 MB and 1,000 pages, with typed blocks.\n - `mistral-document-ai-2512`: Mistral Document AI on Microsoft Foundry, up to 30 MB and 30 pages, with typed blocks.\n - `llamaparse`: LlamaParse's agentic mode, up to 100 MB and 1,000 pages, markdown without blocks.\n - `google-document-ai`: Google Document AI OCR, up to 500 pages, where the deployment has set it up.\n- A URL that isn't one of the app's own files is downloaded first, and can be at most 50 MB. One of the app's own files is read where it sits: up to the model's limit when a model reads it (a PDF, an image), and up to 50 MB for a kind read without a model (text, HTML, spreadsheets and office documents).",
|
|
324
324
|
inputSchema: { "type": "object", "properties": { "url": { "anyOf": [{ "type": "string" }, { "type": "array", "items": { "type": "string" } }] }, "extractionModelOverride": { "type": "object", "properties": { "model": { "type": "string", "description": "Document extraction model identifier" }, "config": { "type": "object", "properties": {}, "required": [], "description": "Additional model-specific configuration" } }, "required": ["model"], "description": "Optional model override. Uses the workspace's default document extraction model if not specified" }, "model": { "type": "string", "description": "Optional extraction model id." }, "includeBlocks": { "type": "boolean", "description": "Add each page's typed blocks, with their bounding boxes, and the page's size to `documents`. Off by default, since they're several times the size of the text." } }, "required": ["url"] },
|
|
325
325
|
outputSchema: { "type": "object", "properties": { "text": { "anyOf": [{ "type": "string" }, { "type": "array", "items": { "type": "string" } }] }, "documents": { "type": "array", "items": { "type": "object", "properties": { "url": { "type": "string", "description": "The URL the document was read from." }, "text": { "type": "string", "description": "The whole document's text." }, "pages": { "type": "array", "items": { "type": "object", "properties": { "pageNumber": { "type": "number", "description": "1-based." }, "markdown": { "type": "string", "description": "The page's text, as markdown." }, "blocks": { "type": "array", "items": { "type": "string" }, "description": "The page's typed regions (title, text, table, image, header, footer\u2026), each with its bounding box, when `includeBlocks` was set and the model returns them." }, "dimensions": { "type": "object", "properties": { "width": { "type": "number" }, "height": { "type": "number" }, "dpi": { "type": "number" } }, "required": ["width", "height"], "description": "The page's size in pixels, the space `blocks`' bounding boxes are in." } }, "required": ["pageNumber", "markdown"], "description": "One page of an extracted document." }, "description": "Its pages, in order. A format read without a model is one page." }, "pagesProcessed": { "type": "number", "description": "The pages the model read and billed, 0 for a format read without one." } }, "required": ["url", "text", "pages", "pagesProcessed"] }, "description": "Each URL's document, in the order given, page by page." } }, "required": ["text", "documents"] }
|
|
326
326
|
},
|
|
@@ -10143,7 +10143,7 @@ async function startMcpServer(options) {
|
|
|
10143
10143
|
capabilities: { tools: {} },
|
|
10144
10144
|
serverInfo: {
|
|
10145
10145
|
name: "mindstudio-agent",
|
|
10146
|
-
version: "0.1.
|
|
10146
|
+
version: "0.1.183"
|
|
10147
10147
|
},
|
|
10148
10148
|
instructions: 'Welcome to MindStudio \u2014 a platform with 200+ AI models, 850+ third-party integrations, and pre-built agents.\n\nGetting started:\n1. Call `ask` with any question about the SDK \u2014 it knows every action, model, and connector and returns working code with real model IDs and config options. Examples: ask("generate an image with FLUX"), ask("what models support vision?"), ask("how do I send a Slack message?").\n2. Call `changeName` to set your display name \u2014 use your name or whatever your user calls you. This is how you\'ll appear in MindStudio request logs.\n3. If you have a profile picture, pass its public image URL to `changeProfilePicture`.\n4. For manual browsing, call `listActions` to discover all available actions.\n\nThen use the tools to generate text, images, video, audio, search the web, work with data sources, run agents, and more.\n\nImportant:\n- AI-powered actions (text generation, image generation, video, audio, etc.) cost money. Before running these, call `estimateActionCost` and confirm with the user before proceeding \u2014 unless they\'ve explicitly told you to go ahead.\n- Not all agents from `listAgents` are configured for API use. Do not try to run an agent just because it appears in the list \u2014 it will likely fail. Only run agents the user specifically asks you to run.'
|
|
10149
10149
|
});
|
|
@@ -11183,7 +11183,7 @@ function isNewerVersion(current, latest) {
|
|
|
11183
11183
|
return false;
|
|
11184
11184
|
}
|
|
11185
11185
|
async function checkForUpdate() {
|
|
11186
|
-
const currentVersion = "0.1.
|
|
11186
|
+
const currentVersion = "0.1.183";
|
|
11187
11187
|
if (!currentVersion || process.env.MINDSTUDIO_NO_UPDATE_CHECK) {
|
|
11188
11188
|
return null;
|
|
11189
11189
|
}
|
|
@@ -11218,7 +11218,7 @@ async function checkForUpdate() {
|
|
|
11218
11218
|
}
|
|
11219
11219
|
}
|
|
11220
11220
|
function printUpdateNotice(latestVersion) {
|
|
11221
|
-
const currentVersion = "0.1.
|
|
11221
|
+
const currentVersion = "0.1.183";
|
|
11222
11222
|
process.stderr.write(
|
|
11223
11223
|
`
|
|
11224
11224
|
${ansi2.cyanBright("Update available")} ${ansi2.gray(currentVersion + " \u2192")} ${ansi2.cyanBold(latestVersion)}
|
|
@@ -11231,7 +11231,7 @@ function isStandaloneBinary() {
|
|
|
11231
11231
|
return !argv1.includes("node_modules");
|
|
11232
11232
|
}
|
|
11233
11233
|
async function cmdUpdate() {
|
|
11234
|
-
const currentVersion = "0.1.
|
|
11234
|
+
const currentVersion = "0.1.183";
|
|
11235
11235
|
process.stderr.write(
|
|
11236
11236
|
` ${ansi2.gray("Current version:")} ${currentVersion}
|
|
11237
11237
|
`
|
|
@@ -11372,7 +11372,7 @@ async function cmdLogin(options) {
|
|
|
11372
11372
|
process.stderr.write("\n");
|
|
11373
11373
|
printLogo();
|
|
11374
11374
|
process.stderr.write("\n");
|
|
11375
|
-
const ver = "0.1.
|
|
11375
|
+
const ver = "0.1.183";
|
|
11376
11376
|
process.stderr.write(
|
|
11377
11377
|
` ${ansi2.bold("MindStudio Agent")} ${ver ? " " + ansi2.gray("v" + ver) : ""}
|
|
11378
11378
|
`
|
|
@@ -11407,7 +11407,7 @@ async function cmdLogin(options) {
|
|
|
11407
11407
|
`Failed to request auth URL: ${authRes.status} ${authRes.statusText}`
|
|
11408
11408
|
);
|
|
11409
11409
|
}
|
|
11410
|
-
const { url, token } = await authRes.json();
|
|
11410
|
+
const { url, token, expiresInSeconds } = await authRes.json();
|
|
11411
11411
|
openBrowser(url);
|
|
11412
11412
|
process.stderr.write(
|
|
11413
11413
|
` ${ansi2.cyanBright("Opening browser to authenticate...")}
|
|
@@ -11418,7 +11418,9 @@ async function cmdLogin(options) {
|
|
|
11418
11418
|
`
|
|
11419
11419
|
);
|
|
11420
11420
|
const POLL_INTERVAL = 2e3;
|
|
11421
|
-
const MAX_ATTEMPTS =
|
|
11421
|
+
const MAX_ATTEMPTS = Math.ceil(
|
|
11422
|
+
(expiresInSeconds ?? 120) * 1e3 / POLL_INTERVAL
|
|
11423
|
+
);
|
|
11422
11424
|
for (let attempt = 0; attempt < MAX_ATTEMPTS; attempt++) {
|
|
11423
11425
|
await sleep5(POLL_INTERVAL);
|
|
11424
11426
|
const frame = SPINNER_FRAMES[attempt % SPINNER_FRAMES.length];
|
|
@@ -11700,7 +11702,7 @@ async function main() {
|
|
|
11700
11702
|
const updatePromise = command !== "mcp" && command !== "login" && command !== "update" ? checkForUpdate() : Promise.resolve(null);
|
|
11701
11703
|
try {
|
|
11702
11704
|
if (command === "version" || command === "-v") {
|
|
11703
|
-
process.stdout.write("0.1.
|
|
11705
|
+
process.stdout.write("0.1.183\n");
|
|
11704
11706
|
return;
|
|
11705
11707
|
}
|
|
11706
11708
|
if (command === "login") {
|
|
@@ -684,7 +684,7 @@ export interface StepMethods {
|
|
|
684
684
|
* - `mistral-document-ai-2512`: Mistral Document AI on Microsoft Foundry, up to 30 MB and 30 pages, with typed blocks.
|
|
685
685
|
* - `llamaparse`: LlamaParse's agentic mode, up to 100 MB and 1,000 pages, markdown without blocks.
|
|
686
686
|
* - `google-document-ai`: Google Document AI OCR, up to 500 pages, where the deployment has set it up.
|
|
687
|
-
* -
|
|
687
|
+
* - A URL that isn't one of the app's own files is downloaded first, and can be at most 50 MB. One of the app's own files is read where it sits: up to the model's limit when a model reads it (a PDF, an image), and up to 50 MB for a kind read without a model (text, HTML, spreadsheets and office documents).
|
|
688
688
|
*
|
|
689
689
|
* @example
|
|
690
690
|
* ```typescript
|
package/dist/index.js
CHANGED
|
@@ -5169,7 +5169,7 @@ var stepMetadata = {
|
|
|
5169
5169
|
"extractText": {
|
|
5170
5170
|
stepType: "extractText",
|
|
5171
5171
|
description: "Download a file from a URL and extract its text content. Supports PDFs (including scanned/image-based PDFs via OCR), plain text files, and other document formats.",
|
|
5172
|
-
usageNotes: "- Best suited for PDFs and raw text/document files. For web pages, use the scrapeUrl step instead.\n- Handles both text-layer PDFs and image-based/scanned PDFs (e.g. Figma/Canva exports, scanned documents). Image-based PDFs are processed with OCR automatically \u2014 there is no need to convert PDF pages to images first.\n- Plain text, code, HTML, spreadsheets and Office documents (docx, pptx, odt, rtf, epub) are read without a model and cost nothing. PDFs and images go to the extraction model, billed per page.\n- Accepts a single URL, a comma-separated list of URLs, or a JSON array of URLs.\n- Nothing is published. One of the app's own files \u2014 including a private one, or a share link \u2014 is read where it sits; any other URL is downloaded and staged privately for the extraction model, then discarded when extraction finishes.\n- `text` is each document's text: a string for one URL, an array for several. `documents` has each document page by page, with its page numbers and the pages billed. Set `includeBlocks` to add each page's typed blocks (title, text, table, image, header, footer\u2026) with their bounding boxes, and the page's size.\n- A file that can't be read (unreachable, an unsupported type, past the model's limits, or refused by the model) fails the step with the reason.\n- `extractionModelOverride.model` picks the model; omit it to use the workspace's default:\n - `mistral-ocr-4.1` (the default): PDFs and images up to 50 MB and 1,000 pages, with typed blocks.\n - `mistral-document-ai-2512`: Mistral Document AI on Microsoft Foundry, up to 30 MB and 30 pages, with typed blocks.\n - `llamaparse`: LlamaParse's agentic mode, up to 100 MB and 1,000 pages, markdown without blocks.\n - `google-document-ai`: Google Document AI OCR, up to 500 pages, where the deployment has set it up.\n-
|
|
5172
|
+
usageNotes: "- Best suited for PDFs and raw text/document files. For web pages, use the scrapeUrl step instead.\n- Handles both text-layer PDFs and image-based/scanned PDFs (e.g. Figma/Canva exports, scanned documents). Image-based PDFs are processed with OCR automatically \u2014 there is no need to convert PDF pages to images first.\n- Plain text, code, HTML, spreadsheets and Office documents (docx, pptx, odt, rtf, epub) are read without a model and cost nothing. PDFs and images go to the extraction model, billed per page.\n- Accepts a single URL, a comma-separated list of URLs, or a JSON array of URLs.\n- Nothing is published. One of the app's own files \u2014 including a private one, or a share link \u2014 is read where it sits; any other URL is downloaded and staged privately for the extraction model, then discarded when extraction finishes.\n- `text` is each document's text: a string for one URL, an array for several. `documents` has each document page by page, with its page numbers and the pages billed. Set `includeBlocks` to add each page's typed blocks (title, text, table, image, header, footer\u2026) with their bounding boxes, and the page's size.\n- A file that can't be read (unreachable, an unsupported type, past the model's limits, or refused by the model) fails the step with the reason.\n- `extractionModelOverride.model` picks the model; omit it to use the workspace's default:\n - `mistral-ocr-4.1` (the default): PDFs and images up to 50 MB and 1,000 pages, with typed blocks.\n - `mistral-document-ai-2512`: Mistral Document AI on Microsoft Foundry, up to 30 MB and 30 pages, with typed blocks.\n - `llamaparse`: LlamaParse's agentic mode, up to 100 MB and 1,000 pages, markdown without blocks.\n - `google-document-ai`: Google Document AI OCR, up to 500 pages, where the deployment has set it up.\n- A URL that isn't one of the app's own files is downloaded first, and can be at most 50 MB. One of the app's own files is read where it sits: up to the model's limit when a model reads it (a PDF, an image), and up to 50 MB for a kind read without a model (text, HTML, spreadsheets and office documents).",
|
|
5173
5173
|
inputSchema: { "type": "object", "properties": { "url": { "anyOf": [{ "type": "string" }, { "type": "array", "items": { "type": "string" } }] }, "extractionModelOverride": { "type": "object", "properties": { "model": { "type": "string", "description": "Document extraction model identifier" }, "config": { "type": "object", "properties": {}, "required": [], "description": "Additional model-specific configuration" } }, "required": ["model"], "description": "Optional model override. Uses the workspace's default document extraction model if not specified" }, "model": { "type": "string", "description": "Optional extraction model id." }, "includeBlocks": { "type": "boolean", "description": "Add each page's typed blocks, with their bounding boxes, and the page's size to `documents`. Off by default, since they're several times the size of the text." } }, "required": ["url"] },
|
|
5174
5174
|
outputSchema: { "type": "object", "properties": { "text": { "anyOf": [{ "type": "string" }, { "type": "array", "items": { "type": "string" } }] }, "documents": { "type": "array", "items": { "type": "object", "properties": { "url": { "type": "string", "description": "The URL the document was read from." }, "text": { "type": "string", "description": "The whole document's text." }, "pages": { "type": "array", "items": { "type": "object", "properties": { "pageNumber": { "type": "number", "description": "1-based." }, "markdown": { "type": "string", "description": "The page's text, as markdown." }, "blocks": { "type": "array", "items": { "type": "string" }, "description": "The page's typed regions (title, text, table, image, header, footer\u2026), each with its bounding box, when `includeBlocks` was set and the model returns them." }, "dimensions": { "type": "object", "properties": { "width": { "type": "number" }, "height": { "type": "number" }, "dpi": { "type": "number" } }, "required": ["width", "height"], "description": "The page's size in pixels, the space `blocks`' bounding boxes are in." } }, "required": ["pageNumber", "markdown"], "description": "One page of an extracted document." }, "description": "Its pages, in order. A format read without a model is one page." }, "pagesProcessed": { "type": "number", "description": "The pages the model read and billed, 0 for a format read without one." } }, "required": ["url", "text", "pages", "pagesProcessed"] }, "description": "Each URL's document, in the order given, page by page." } }, "required": ["text", "documents"] }
|
|
5175
5175
|
},
|