@youdie006/prodex 0.19.2 → 0.21.0

package/README.md

@@ -35,7 +35,7 @@ prodex ask --file src/auth.ts "Review this for security holes"
 
 `prodex ask` is the short form of `prodex pro browser ask`; the full form and every flag work identically. In an interactive terminal, `login` keeps watching the opened window and tells you exactly which manual step is still missing (log in, clear a check, open a chat) until it reports READY. If you skip `login` and the browser is not running, an interactive `ask` recovers on its own: it launches the dedicated browser, waits for your saved session to be READY, and retries the send once (disable with `--no-auto-login`; scripts opt in with `--auto-login`). While ChatGPT thinks, `prodex` prints progress to stderr (connecting, prompt sent, elapsed seconds while generating), so a multi-minute Pro answer never looks frozen.
 
-The answer prints to your terminal and is saved under `.bridge/` for later (`prodex pro latest` re-prints it). Add `--new-chat` to send into a fresh chat (recommended for repeated consults - long threads eventually confuse send detection). For a structured second-opinion debate between your coding agent and GPT Pro, `prodex pro debate-prompt --topic "..."` prints a ready-to-paste orchestration prompt. `prodex` drives the picker you can see and will not send into a tab it cannot read, so leave the dedicated window on a ChatGPT tab; it sends quietly in the background without stealing focus. Prefer no window at all? See [virtual display](#no-window-at-all-virtual-display-recommended). Pin per-repo defaults once - `prodex setup --model Pro --project "your-project"` - so every ask runs Pro (20-minute timeout) inside that project instead of whatever the ChatGPT UI last had selected; list exact sidebar project names with `prodex pro browser projects`. Pass `--file` more than once to attach several files. When the thread is still generating a previous answer (common right after a timed-out Pro send), the send automatically queues behind it up to the timeout budget; tune that with `--busy-wait-ms` (0 fails fast with a `response_in_progress` blocker). See [First Pro Login](#first-pro-login) for the full flow, and the [FAQ](#faq) if a send stops.
+The answer prints to your terminal and is saved under `.bridge/` for later (`prodex pro latest` re-prints it). Add `--new-chat` to send into a fresh chat (recommended for repeated consults - long threads eventually confuse send detection). For a structured second-opinion debate between your coding agent and GPT Pro, `prodex pro debate-prompt --topic "..."` prints a ready-to-paste orchestration prompt. `prodex` drives the picker you can see and will not send into a tab it cannot read, so leave the dedicated window on a ChatGPT tab; it sends quietly in the background without stealing focus. Prefer no window at all? See [virtual display](#no-window-at-all-virtual-display-recommended). Pin per-repo defaults once - `prodex setup --model Pro --project "your-project"` - so every ask runs Pro (20-minute timeout) inside that project instead of whatever the ChatGPT UI last had selected; list exact sidebar project names with `prodex pro browser projects`. Pass `--file` more than once to inline several files. `--file` puts a text file's CONTENTS into the prompt; `--attach` uploads the file itself, which is the only way to hand ChatGPT a pdf, pptx, xlsx or image and let it parse the original (`prodex ask --attach deck.pptx "Review slides 40-60"`). Both are restricted to paths inside the repo, so an agent cannot upload `~/.ssh` by asking nicely. The upload happens before the prompt is submitted and prodex waits for ChatGPT to finish accepting the file - the browser process reads the path, so the file has to live on the machine running the browser. `--tool` turns on a ChatGPT composer tool for that send: `--tool deep-research` (a browsed report - the timeout rises to 30 minutes automatically, and ChatGPT often replies with a clarifying question first, which you answer with a normal follow-up in the same thread), `--tool web-search` (current facts with sources), `--tool create-image`. Any other label the menu shows works too, so a tool ChatGPT adds later needs no prodex release. When the thread is still generating a previous answer (common right after a timed-out Pro send), the send automatically queues behind it up to the timeout budget; tune that with `--busy-wait-ms` (0 fails fast with a `response_in_progress` blocker). See [First Pro Login](#first-pro-login) for the full flow, and the [FAQ](#faq) if a send stops.
 
 ## Core Shape
 

package/dist/chatgpt-browser.js

@@ -1065,6 +1065,81 @@ async function verifiedClickAt(cdp, x, y, label) {
     await cdp.send("Input.dispatchMouseEvent", { type: "mousePressed", x, y, button: "left", clickCount: 1 });
     await cdp.send("Input.dispatchMouseEvent", { type: "mouseReleased", x, y, button: "left", clickCount: 1 });
 }
+/**
+ * Whether the model picker BUTTON already advertises the requested model, so
+ * the menu never has to be opened. ChatGPT moved the models behind a "Model"
+ * submenu (the top level is now Advanced / Model / Effort), which broke the
+ * flat radio lookup with "Pro option not found in the model menu" - while the
+ * button itself read "Pro, 5 of 5". Selecting what is already selected is
+ * pointless work that a UI change can only break.
+ */
+export function modelButtonAlreadyShows(requestedModel, buttonLabel) {
+    if (!requestedModel || !buttonLabel)
+        return false;
+    const wanted = requestedModel.trim().toLowerCase();
+    if (!wanted)
+        return false;
+    // The label carries decoration ("Pro, 5 of 5.", "GPT-5.6 Pro"), so match the
+    // model name as a WORD inside it rather than by equality.
+    const label = buttonLabel.trim().toLowerCase();
+    const escaped = wanted.replace(/[.*+?^${}()|[\]\\]/g, "\\$&");
+    return new RegExp(`(^|[^a-z0-9])${escaped}([^a-z0-9]|$)`).test(label);
+}
+// ---------------------------------------------------------------------------
+// Power slider (model + effort)
+//
+// ChatGPT replaced the model radio list with ONE slider. Measured live, its
+// five positions render on GPT-5.6 Sol as:
+//   0 Instant · 1 Medium · 2 High · 3 Extra High · 4 Pro
+// so "Pro" is now the top EFFORT, not a model - which is exactly why looking
+// for a "Pro" radio started failing. The slider takes focus and responds to
+// Arrow keys, so selection is: focus, step toward the wanted label, stop.
+// ---------------------------------------------------------------------------
+const POWER_LABEL_SYNONYMS = [
+    { canonical: "instant", aliases: ["instant", "즉시", "빠름", "fast"] },
+    { canonical: "medium", aliases: ["medium", "중간", "보통"] },
+    { canonical: "high", aliases: ["high", "높음"] },
+    { canonical: "extra high", aliases: ["extra high", "extrahigh", "very high", "매우 높음", "매우높음"] },
+    { canonical: "pro", aliases: ["pro", "프로"] }
+];
+function canonicalPowerLabel(value) {
+    const normalized = value.trim().toLowerCase().replace(/\s+/g, " ");
+    const hit = POWER_LABEL_SYNONYMS.find((entry) => entry.aliases.includes(normalized));
+    return hit ? hit.canonical : normalized;
+}
+/** Whether a requested model/effort names the same step the slider renders. */
+export function powerLabelMatches(requested, rendered) {
+    if (!requested || !rendered)
+        return false;
+    return canonicalPowerLabel(requested) === canonicalPowerLabel(rendered);
+}
+/** Slider position plus the Model/Effort readout next to it. */
+export function powerSliderStateExpression() {
+    return `(() => {
+    const slider = document.querySelector('[role="slider"]');
+    const menu = document.querySelector('[data-testid="composer-intelligence-picker-content"]');
+    const lines = menu ? (menu.innerText || "").split(String.fromCharCode(10)).map((l) => l.trim()).filter(Boolean) : [];
+    const after = (label) => { const i = lines.indexOf(label); return i >= 0 ? lines[i + 1] : null; };
+    if (!slider) return { ok: false, reason: "power slider not found", lines };
+    return {
+      ok: true,
+      position: Number(slider.getAttribute("aria-valuenow")),
+      min: Number(slider.getAttribute("aria-valuemin")),
+      max: Number(slider.getAttribute("aria-valuemax")),
+      model: after("Model"),
+      effort: after("Effort"),
+      lines
+    };
+  })()`;
+}
+export function focusPowerSliderExpression() {
+    return `(() => {
+    const slider = document.querySelector('[role="slider"]');
+    if (!slider) return { ok: false, reason: "power slider not found" };
+    slider.focus();
+    return { ok: document.activeElement === slider };
+  })()`;
+}
 export function modelButtonRectExpression() {
     return `(() => {${CLICK_POINT_SNIPPET}
     const c = document.querySelector('#prompt-textarea,[contenteditable="true"],textarea');
@@ -1075,7 +1150,10 @@ export function modelButtonRectExpression() {
       return /\\S/.test(t) && !/파일|첨부|받아쓰기|음성|dictation|attach|file|voice|record|search|mic/i.test(t + aria);
     });
     if (!b) return { ok: false, reason: "model selector button not found" };
-    return clickPoint(b);
+    // Return the label too: the caller compares it against the requested model
+    // to skip opening the menu when it is already selected.
+    const label = ((b.getAttribute("aria-label") || b.textContent || "").trim().split(String.fromCharCode(10))[0] || "").trim();
+    return { ...clickPoint(b), label };
   })()`;
 }
 export function menuItemRectExpression(label) {
@@ -1211,7 +1289,44 @@ async function assertSelectionCommitted(cdp, label) {
         throw new Error(`ChatGPT selection "${label}" did not commit; the model menu stayed open. Retry, or pick it manually in the visible browser.`);
     }
 }
-async function selectModelReasoning(cdp, options) {
+/**
+ * Move the power slider until its Effort readout is the requested step. The
+ * menu must already be open. Returns the quota line so the caller can warn
+ * when Pro runs are nearly spent.
+ */
+async function selectPowerStep(cdp, requested) {
+    const focused = await cdp.evaluate(focusPowerSliderExpression());
+    if (!focused?.ok) {
+        throw new Error(focused?.reason ??
+            "ChatGPT's model picker did not expose its power slider, so the requested model/effort could not be selected.");
+    }
+    let state = await cdp.evaluate(powerSliderStateExpression());
+    if (!state?.ok)
+        throw new Error(state?.reason ?? "Could not read ChatGPT's power slider");
+    const steps = (state.max ?? 4) - (state.min ?? 0) + 1;
+    for (let attempt = 0; attempt <= steps * 2; attempt += 1) {
+        if (state.effort && powerLabelMatches(requested, state.effort))
+            return { effort: state.effort };
+        // Walk upward first, then back down: the labels are ordered, but their
+        // exact set can change, so this never assumes a fixed index for a name.
+        const atTop = (state.position ?? 0) >= (state.max ?? 4);
+        const key = attempt < steps && !atTop ? "ArrowRight" : "ArrowLeft";
+        await dispatchArrowKey(cdp, key);
+        await sleep(400);
+        state = await cdp.evaluate(powerSliderStateExpression());
+        if (!state?.ok)
+            throw new Error(state?.reason ?? "Could not read ChatGPT's power slider");
+    }
+    const available = (state.lines ?? []).join(" / ");
+    throw new Error(`ChatGPT's model picker has no "${requested}" step. It showed: ${available}`);
+}
+async function dispatchArrowKey(cdp, key) {
+    const code = key;
+    const virtualKey = key === "ArrowLeft" ? 37 : 39;
+    await cdp.send("Input.dispatchKeyEvent", { type: "rawKeyDown", key, code, windowsVirtualKeyCode: virtualKey });
+    await cdp.send("Input.dispatchKeyEvent", { type: "keyUp", key, code, windowsVirtualKeyCode: virtualKey });
+}
+async function selectModelReasoning(cdp, options, selectionWarnings = []) {
     if (!options.model && !options.proMode && !options.effort)
         return;
     // --pro-mode selects a Pro sub-mode, so it is meaningless with a non-Pro
@@ -1236,6 +1351,13 @@ async function selectModelReasoning(cdp, options) {
     if (!button.ok || button.x === undefined || button.y === undefined) {
         throw new Error(button.reason ?? "Could not open the ChatGPT model selector");
     }
+    // Skip the menu entirely when the picker already shows the requested model:
+    // it is the same end state, and it survives ChatGPT reshuffling the menu
+    // (which it did - the models moved behind a "Model" submenu and every
+    // --model Pro send started failing).
+    if (options.model && !options.proMode && !options.effort && modelButtonAlreadyShows(options.model, button.label)) {
+        return;
+    }
     try {
         // The hover-verified click can be transiently refused right after a page
         // transition (measured live: the just-closed create-project modal's
@@ -1260,6 +1382,19 @@ async function selectModelReasoning(cdp, options) {
         const opened = await waitForExpressionTrue(cdp, menuOpenExpression(), MENU_OPEN_TIMEOUT_MS);
         if (!opened)
             throw new Error("ChatGPT model menu did not open after clicking the selector");
+        // Current ChatGPT: one power slider (Instant/Medium/High/Extra High/Pro on
+        // GPT-5.6 Sol) instead of a model radio list, so "Pro" is the top EFFORT.
+        // Drive it when it is there and fall through to the legacy radio path when
+        // it is not, so both UI generations work.
+        const sliderState = await cdp.evaluate(powerSliderStateExpression());
+        if (sliderState?.ok) {
+            const wanted = options.effort ?? options.model;
+            if (wanted) {
+                await selectPowerStep(cdp, wanted);
+            }
+            await dispatchEscapeKey(cdp);
+            return;
+        }
         const wantsProMode = Boolean(options.proMode) && (!options.model || /pro/i.test(options.model));
         if (wantsProMode && options.proMode) {
             // Open the Pro sub-mode submenu via the chevron, then pick 기본/확장.
@@ -1709,7 +1844,7 @@ export async function sendChatGptPrompt(options) {
         await cdp.send("Runtime.enable");
         await selectProject(cdp, options);
         try {
-            await selectModelReasoning(cdp, options);
+            await selectModelReasoning(cdp, options, sendWarnings);
         }
         catch (modelError) {
             // Pro sub-mode isn't exposed in this UI yet (staged rollout). Pro itself is
@@ -1729,7 +1864,17 @@ export async function sendChatGptPrompt(options) {
         // into; a --project/--project-new hop lands on a page with its own counts.
         beforeSubmit = await evaluateOnPage(page, answerExpression());
         dbgSend(`baseline url=${beforeSubmit.url} user=${beforeSubmit.userMessageCount} assistant=${beforeSubmit.assistantMessageCount}`);
-        await insertComposerTextViaCdp(cdp, options.prompt, page);
+        // Attach BEFORE typing: the upload is the slow part, and a file that
+        // arrives after the prompt is submitted is a file ChatGPT never saw.
+        if (options.attachments && options.attachments.length > 0) {
+            emitProgress("selecting", `uploading ${options.attachments.length} file(s)`);
+            const uploaded = await attachFilesToComposer(cdp, options.attachments);
+            emitProgress("selecting", `attached ${uploaded.attached.join(", ")}`);
+        }
+        const toolLabels = (options.tools ?? []).map(resolveComposerToolLabel);
+        if (toolLabels.length > 0)
+            emitProgress("selecting", `tools=${toolLabels.join(", ")}`);
+        await insertComposerTextViaCdp(cdp, options.prompt, page, toolLabels);
         // The send button renders asynchronously after the prompt lands. Poll for it
         // BEFORE submitting so (a) submitButtonFound reflects whether the control
         // actually EXISTS - otherwise a successful Enter-key submit skips the fallback
@@ -1738,7 +1883,12 @@ export async function sendChatGptPrompt(options) {
         // live: submitExpression finds data-testid="send-button" fine, yet the
         // timeout error blamed a UI change) - and (b) we never press Enter before the
         // composer is submit-ready.
-        submitButtonFound = await waitForExpressionTrue(cdp, `(${submitExpression()}).ok === true`, 3_000);
+        // With an attachment, ChatGPT keeps the send control disabled while it
+        // ingests the file server-side; 3s expired mid-ingest and the send never
+        // posted (measured live on a markdown attachment: the button was enabled
+        // and clickable a moment after prodex gave up).
+        const submitReadyBudgetMs = options.attachments && options.attachments.length > 0 ? 120_000 : 3_000;
+        submitButtonFound = await waitForExpressionTrue(cdp, `(${submitExpression()}).ok === true`, submitReadyBudgetMs);
         // Submit. Prefer the Enter key: it goes to the focused composer and does
         // not depend on coordinates, whereas the send button moves ~100px as the
         // composer grows after the prompt lands, so a click at captured coordinates
@@ -2293,13 +2443,191 @@ export function insertComposerTextInPageExpression(text) {
     return { ok: true };
   })()`;
 }
-export function composerTextStateExpression(expectedText) {
+// ---------------------------------------------------------------------------
+// Attachments (real file upload)
+//
+// `--file` inlines a file's TEXT into the prompt; this uploads the file itself
+// through the composer's file input, which is the only way to hand ChatGPT a
+// pdf/pptx/image and let it parse the original. CDP's DOM.setFileInputFiles
+// sets the input without any file dialog.
+// ---------------------------------------------------------------------------
+/**
+ * The composer's general file input. Measured live: ChatGPT renders three
+ * inputs - one general (accept="") plus two accept="image/*" - and attaching a
+ * document to an image-only input silently does nothing.
+ */
+export function composerFileInputSelector() {
+    return 'input[type="file"]:not([accept*="image"])';
+}
+/**
+ * Which of the expected attachments the composer shows, and whether one is
+ * still uploading.
+ *
+ * Measured live: a DOCUMENT chip carries its filename in visible text, but an
+ * IMAGE renders as a blob: thumbnail whose only filename is on the remove
+ * button's aria-label ("Remove file 1: cli-banner.png"). Reading innerText
+ * alone therefore reported a perfectly good image upload as never accepted.
+ */
+export function attachmentStateExpression(fileNames) {
+    const namesJson = JSON.stringify(fileNames);
+    return `(() => {
+    const names = ${namesJson};
+    const text = document.body ? document.body.innerText || "" : "";
+    const labels = [...document.querySelectorAll('[aria-label],[title],[alt]')]
+      .map((el) => (el.getAttribute("aria-label") || el.getAttribute("title") || el.getAttribute("alt") || ""))
+      .join(String.fromCharCode(10));
+    const haystack = text + String.fromCharCode(10) + labels;
+    const present = names.filter((name) => haystack.includes(name));
+    const attachedFiles = [...document.querySelectorAll('input[type="file"]')]
+      .reduce((total, el) => total + (el.files ? el.files.length : 0), 0);
+    // A visible progressbar means a file is still going up; sending now would
+    // post the prompt without it.
+    const uploading = document.querySelectorAll('[role="progressbar"]').length > 0;
+    return { ok: true, present, attachedFiles, uploading };
+  })()`;
+}
+/** How many attachments are already sitting in the composer (read-only). */
+export function attachmentPresenceExpression() {
+    return `(() => {
+    const buttons = [...document.querySelectorAll('button[aria-label]')].filter((b) =>
+      /remove file|파일 제거|첨부 제거/i.test(b.getAttribute("aria-label") || "")
+    );
+    return { ok: true, removed: buttons.length };
+  })()`;
+}
+/**
+ * Attach files to the composer and wait until ChatGPT has taken them. The
+ * files must be readable by the BROWSER process, so a browser on another
+ * machine (or another container) cannot see the caller's paths - that shows up
+ * as the attachment never appearing, which this reports as such.
+ */
+export async function attachFilesToComposer(cdp, absolutePaths, options = {}) {
+    if (absolutePaths.length === 0)
+        return { attached: [] };
+    await cdp.send("DOM.enable");
+    // Drop anything a previous (failed) send left attached: it would ride along
+    // with this prompt, which is exactly the kind of silent contamination the
+    // composer text check already guards against. This must happen BEFORE the
+    // input node is resolved - removing a chip re-renders the composer and
+    // replaces the input element, and setting files on the old (detached) node
+    // silently does nothing (measured live: the file input held the file while
+    // the UI showed no attachment at all).
+    // Leftover attachments from a previous (failed) send would ride along with
+    // this prompt. Clearing them by clicking the remove buttons WEDGES the
+    // composer: measured live, after a removal the file input accepts files
+    // (input.files becomes 1) while the UI never renders the chip again, and
+    // every later attach in that tab silently does nothing. A reload is the only
+    // reliable reset, and it costs a few seconds only when there is something to
+    // clear.
+    const stale = await cdp.evaluate(attachmentPresenceExpression());
+    if ((stale?.removed ?? 0) > 0) {
+        await cdp.evaluate("location.reload()");
+        await sleep(6_000);
+        const settleDeadline = Date.now() + 15_000;
+        for (;;) {
+            const ready = await cdp.evaluate(composerTextStateExpression());
+            if (ready?.ok || Date.now() >= settleDeadline)
+                break;
+            await sleep(500);
+        }
+    }
+    const document = await cdp.send("DOM.getDocument", { depth: -1, pierce: true });
+    const rootNodeId = document.result?.root?.nodeId;
+    if (rootNodeId === undefined)
+        throw new Error("Could not read the ChatGPT page DOM to attach files.");
+    const input = await cdp.send("DOM.querySelector", { nodeId: rootNodeId, selector: composerFileInputSelector() });
+    const inputNodeId = input.result?.nodeId;
+    if (!inputNodeId) {
+        throw new Error("The ChatGPT composer has no file input to attach to. Open a normal chat (not a shared or read-only view) and retry.");
+    }
+    await cdp.send("DOM.setFileInputFiles", { files: absolutePaths, nodeId: inputNodeId });
+    const fileNames = absolutePaths.map((file) => path.basename(file));
+    const deadline = Date.now() + (options.timeoutMs ?? 120_000);
+    let lastPresent = [];
+    while (Date.now() < deadline) {
+        await sleep(1_000);
+        const state = await cdp.evaluate(attachmentStateExpression(fileNames));
+        lastPresent = state?.present ?? [];
+        if (lastPresent.length === fileNames.length && !state?.uploading)
+            return { attached: lastPresent };
+    }
+    const missing = fileNames.filter((name) => !lastPresent.includes(name));
+    throw new Error(`ChatGPT did not finish accepting ${missing.length > 0 ? missing.join(", ") : fileNames.join(", ")} within the upload budget. The file must be readable by the browser process (same machine), and ChatGPT enforces its own size and type limits.`);
+}
+// ---------------------------------------------------------------------------
+// Composer tools (Deep research, Web search, Create image, connectors)
+//
+// Measured live: a tool is NOT a chip beside the composer - selecting it
+// inserts its name as a token INSIDE the ProseMirror editor, and it survives a
+// page reload. Two consequences drive this code: the tool must be enabled
+// AFTER the composer is cleared (clearing removes it), and the composer text
+// check has to ignore the token or it reads as leftover contamination.
+// ---------------------------------------------------------------------------
+const COMPOSER_TOOL_ALIASES = [
+    { label: "Deep research", aliases: ["deep research", "deep-research", "deepresearch", "deep", "research"] },
+    { label: "Web search", aliases: ["web search", "web-search", "websearch", "search", "web"] },
+    { label: "Create image", aliases: ["create image", "create-image", "image", "img"] }
+];
+/**
+ * Map what a caller typed to the label ChatGPT renders. An unknown value is
+ * passed through unchanged, so a tool ChatGPT adds tomorrow is reachable by
+ * its label without a prodex release.
+ */
+export function resolveComposerToolLabel(requested) {
+    const normalized = requested.trim().toLowerCase();
+    const known = COMPOSER_TOOL_ALIASES.find((tool) => tool.aliases.includes(normalized) || tool.label.toLowerCase() === normalized);
+    return known ? known.label : requested.trim();
+}
+export const DEEP_RESEARCH_TOOL_LABEL = "Deep research";
+const DEEP_RESEARCH_MIN_TIMEOUT_MS = 1_800_000;
+/** Deep research browses for minutes; the ordinary budget abandons it mid-report. */
+export function defaultTimeoutForTools(tools, fallbackMs) {
+    const wantsDeepResearch = tools.some((tool) => resolveComposerToolLabel(tool) === DEEP_RESEARCH_TOOL_LABEL);
+    return wantsDeepResearch ? Math.max(fallbackMs, DEEP_RESEARCH_MIN_TIMEOUT_MS) : fallbackMs;
+}
+export function composerToolsButtonRectExpression() {
+    return `(() => {${CLICK_POINT_SNIPPET}
+    const b = document.querySelector('[data-testid="composer-plus-btn"]');
+    if (!b) return { ok: false, reason: "composer tools button not found" };
+    return clickPoint(b);
+  })()`;
+}
+/** Click point for a tools-menu entry, matched by its visible label. */
+export function composerToolEntryRectExpression(label) {
+    const labelJson = JSON.stringify(label);
+    return `(() => {${CLICK_POINT_SNIPPET}
+    const wanted = ${labelJson}.trim().toLowerCase();
+    const leaves = [...document.querySelectorAll("div,span,button,a")].filter((el) => el.children.length === 0);
+    const leaf = leaves.find((el) => (el.textContent || "").trim().toLowerCase() === wanted);
+    if (!leaf) {
+      const available = [...new Set(leaves.map((el) => (el.textContent || "").trim()).filter((t) => t.length > 1 && t.length < 30))].slice(0, 20);
+      return { ok: false, reason: "tool not found in the composer tools menu", available };
+    }
+    const target = leaf.closest('[role="menuitem"],[role="option"],button,a') || leaf.parentElement || leaf;
+    return clickPoint(target);
+  })()`;
+}
+/** Tool tokens currently sitting in the composer. */
+export function activeComposerToolsExpression(labels) {
+    const labelsJson = JSON.stringify(labels);
+    return `(() => {
+    const el = document.querySelector('#prompt-textarea,[contenteditable="true"]');
+    const text = el ? (el.innerText || "") : "";
+    return { ok: true, active: ${labelsJson}.filter((label) => text.includes(label)) };
+  })()`;
+}
+export function composerTextStateExpression(expectedText, toolLabels = []) {
     const expectedJson = JSON.stringify(expectedText ?? null);
+    const toolLabelsJson = JSON.stringify(toolLabels);
     return `(() => {
     ${composerExpressionHelpers()}
     const el = findChatGptComposerCandidate();
     if (!el) return { ok: false, reason: "No visible composer" };
-    const raw = ("value" in el ? el.value : el.innerText || el.textContent || "").trim();
+    let raw = ("value" in el ? el.value : el.innerText || el.textContent || "").trim();
+    // An enabled tool lives INSIDE the composer as a token; it is not leftover
+    // text, so strip it before comparing against the prompt.
+    for (const label of ${toolLabelsJson}) raw = raw.split(label).join(" ");
+    raw = raw.trim();
     if (!raw) return { ok: false, reason: "Composer stayed empty after text insertion" };
     const expected = ${expectedJson};
     if (expected === null) return { ok: true, actualText: raw.slice(0, 120) };
@@ -2313,7 +2641,56 @@ export function composerTextStateExpression(expectedText) {
 // Focus the composer, clear any leftover text submit-safely, type the prompt
 // with native CDP input so ProseMirror registers it, then verify the composer
 // holds exactly the prompt.
-async function insertComposerTextViaCdp(cdp, text, page) {
+/**
+ * Turn on a composer tool by its menu label and confirm the token landed in
+ * the composer. Called after the composer is cleared and before the prompt is
+ * typed, because clearing the composer removes the token.
+ */
+export async function enableComposerTools(cdp, labels) {
+    const enabled = [];
+    for (const label of labels) {
+        const already = await cdp.evaluate(activeComposerToolsExpression([label]));
+        if ((already?.active ?? []).includes(label)) {
+            enabled.push(label);
+            continue;
+        }
+        const button = await cdp.evaluate(composerToolsButtonRectExpression());
+        if (!button.ok || button.x === undefined || button.y === undefined) {
+            throw new Error(button.reason ?? "Could not open the ChatGPT composer tools menu");
+        }
+        await dispatchMouseClickAt(cdp, button.x, button.y);
+        let entry = { ok: false };
+        const menuDeadline = Date.now() + 6_000;
+        for (;;) {
+            entry = await cdp.evaluate(composerToolEntryRectExpression(label));
+            if (entry.ok && entry.x !== undefined && entry.y !== undefined)
+                break;
+            if (Date.now() >= menuDeadline)
+                break;
+            await sleep(250);
+        }
+        if (!entry.ok || entry.x === undefined || entry.y === undefined) {
+            await dispatchEscapeKey(cdp);
+            const available = entry.available?.length ? ` Menu showed: ${entry.available.slice(0, 12).join(", ")}.` : "";
+            throw new Error(`ChatGPT's composer tools menu has no "${label}".${available}`);
+        }
+        await dispatchMouseClickAt(cdp, entry.x, entry.y);
+        const activeDeadline = Date.now() + 8_000;
+        let active = false;
+        for (;;) {
+            const state = await cdp.evaluate(activeComposerToolsExpression([label]));
+            active = (state?.active ?? []).includes(label);
+            if (active || Date.now() >= activeDeadline)
+                break;
+            await sleep(250);
+        }
+        if (!active)
+            throw new Error(`Selected "${label}" but the composer never showed it as active.`);
+        enabled.push(label);
+    }
+    return enabled;
+}
+async function insertComposerTextViaCdp(cdp, text, page, toolLabels = []) {
     const prepared = await cdp.evaluate(prepareComposerExpression());
     if (!prepared.ok)
         throw new Error(prepared.reason ?? "Could not focus the ChatGPT composer");
@@ -2329,6 +2706,12 @@ async function insertComposerTextViaCdp(cdp, text, page) {
         await cdp.send("Input.dispatchKeyEvent", { type: "keyUp", key: "Backspace", code: "Backspace", windowsVirtualKeyCode: 8 });
         await sleep(100);
     }
+    // The composer is clear at this point, so the tool tokens can go in now -
+    // enabling them earlier would have been wiped by the clear above.
+    if (toolLabels.length > 0) {
+        await enableComposerTools(cdp, toolLabels);
+        await sleep(300);
+    }
     // Insertion path, chosen by size:
     //
     // Short prompts go through Input.insertText - real key-level input events,
@@ -2367,7 +2750,7 @@ async function insertComposerTextViaCdp(cdp, text, page) {
         }
     }
     await sleep(200);
-    const state = await cdp.evaluate(composerTextStateExpression(text));
+    const state = await cdp.evaluate(composerTextStateExpression(text, toolLabels));
     if (!state.ok)
         throw new Error(state.reason ?? "Composer stayed empty after text insertion");
 }

package/dist/cli-args.js

@@ -261,6 +261,10 @@ export const ASK_PRO_SELECTION_VALUE_FLAGS = ["--project", "--project-new", "--m
 export const ASK_PRO_VALUE_FLAGS = new Set([
     "--cwd",
     "--file",
+    // Upload the file itself (pdf/pptx/image) instead of inlining its text.
+    "--attach",
+    // Composer tools: deep-research, web-search, create-image, ...
+    "--tool",
     "--port",
     "--timeout-ms",
     "--busy-wait-ms",

package/dist/cli-help.js

@@ -17,7 +17,7 @@ First-time setup:
 
 Ask / consult commands:
   prodex ask [same flags as pro browser ask] "prompt"  # top-level shortcut for pro browser ask
-  prodex pro ask [--dry-run] [--cwd /absolute/path/to/repo] [--file path] "prompt"  # dry-run preview
+  prodex pro ask [--dry-run] [--cwd /absolute/path/to/repo] [--file path] [--attach path] [--tool deep-research|web-search|create-image] "prompt"  # dry-run preview
   prodex pro debate-prompt [--topic "..."] [--rounds 2] [--source-cli /absolute/path/to/dist/cli.js]  # print an agent prompt for a structured GPT Pro debate
   prodex pro browser login [--cwd /absolute/path/to/repo] [--dry-run] [--source-cli /absolute/path/to/dist/cli.js] [--profile-dir path] [--port 9333] [--url https://chatgpt.com/...] [--launch-timeout-ms 5000] [--wait|--no-wait] [--headless|--minimized|--virtual-display] [--wait-timeout-ms 300000]  # preview/open visible browser login
   prodex pro browser help [--source-cli /absolute/path/to/dist/cli.js]
@@ -26,14 +26,14 @@ Ask / consult commands:
   prodex pro browser models [--source-cli /absolute/path/to/dist/cli.js] [--port 9333] [--timeout-ms 15000]  # read-only list of model menu options
   prodex pro browser projects [--source-cli /absolute/path/to/dist/cli.js] [--port 9333] [--timeout-ms 15000]  # read-only list of sidebar project names (for --project)
   prodex pro browser recover [--source-cli /absolute/path/to/dist/cli.js] [--cwd /absolute/path/to/repo] [--port 9333] --target-url <thread-url> [--timeout-ms 60000]  # recover a finished answer from a thread whose send timed out
-  prodex pro browser ask [--source-cli /absolute/path/to/dist/cli.js] [--cwd /absolute/path/to/repo] [--port 9333] [--timeout-ms 300000] [--busy-wait-ms 600000] [--target-url url --confirm-target] [--new-chat] [--stdin] [--json] [--auto-login|--no-auto-login] [--file path] [--model Pro] [--pro-mode 기본|확장] [--effort 즉시|중간|높음|"매우 높음"] [--project "name" | --project-new "name"] "prompt"  # explicit visible-browser send
+  prodex pro browser ask [--source-cli /absolute/path/to/dist/cli.js] [--cwd /absolute/path/to/repo] [--port 9333] [--timeout-ms 300000] [--busy-wait-ms 600000] [--target-url url --confirm-target] [--new-chat] [--stdin] [--json] [--auto-login|--no-auto-login] [--file path] [--attach path] [--tool deep-research|web-search|create-image] [--model Pro] [--pro-mode 기본|확장] [--effort 즉시|중간|높음|"매우 높음"] [--project "name" | --project-new "name"] "prompt"  # explicit visible-browser send
   prodex pro latest [--source-cli /absolute/path/to/dist/cli.js] [--cwd /absolute/path/to/repo]
   prodex pro list [--source-cli /absolute/path/to/dist/cli.js] [--cwd /absolute/path/to/repo] [--json]
   prodex pro show <task-id|latest> [--source-cli /absolute/path/to/dist/cli.js] [--cwd /absolute/path/to/repo]
 
 Bridge ledger (durable tasks/results/receipts/sessions under .bridge/):
   prodex init [--cwd /absolute/path/to/repo]
-  prodex tasks create [--cwd /absolute/path/to/repo] --title "Title" --prompt "Prompt" [--repo-id id] [--file path]
+  prodex tasks create [--cwd /absolute/path/to/repo] --title "Title" --prompt "Prompt" [--repo-id id] [--file path] [--attach path] [--tool deep-research|web-search|create-image]
   prodex tasks list [--status new|claimed|done|blocked] [--cwd /absolute/path/to/repo] [--json]
   prodex tasks show <task-id|latest> [--cwd /absolute/path/to/repo]
   prodex tasks claim <task-id> [--cwd /absolute/path/to/repo] [--by codex]
@@ -159,14 +159,14 @@ export function printProHelp(stdout) {
     stdout(`prodex pro
 
 Commands:
-  prodex pro ask [--dry-run] [--cwd /absolute/path/to/repo] [--file path] "prompt"
+  prodex pro ask [--dry-run] [--cwd /absolute/path/to/repo] [--file path] [--attach path] [--tool deep-research|web-search|create-image] "prompt"
   prodex pro debate-prompt [--topic "..."] [--rounds 2] [--source-cli /absolute/path/to/dist/cli.js]
   prodex pro browser help [--source-cli /absolute/path/to/dist/cli.js]
   prodex pro browser login [--cwd /absolute/path/to/repo] [--dry-run] [--source-cli /absolute/path/to/dist/cli.js] [--launch-timeout-ms 5000] [--wait|--no-wait] [--headless|--minimized|--virtual-display] [--wait-timeout-ms 300000]
   prodex pro browser check [--source-cli /absolute/path/to/dist/cli.js] [--cwd /absolute/path/to/repo]
   prodex pro browser smoke [--source-cli /absolute/path/to/dist/cli.js] [--cwd /absolute/path/to/repo]
   prodex pro browser models [--source-cli /absolute/path/to/dist/cli.js]
-  prodex pro browser ask [--source-cli /absolute/path/to/dist/cli.js] [--cwd /absolute/path/to/repo] [--target-url url --confirm-target] [--new-chat] [--stdin] [--json] [--auto-login|--no-auto-login] [--file path] [--model Pro] [--pro-mode 기본|확장] [--effort 즉시|중간|높음|"매우 높음"] [--project "name" | --project-new "name"] "prompt"
+  prodex pro browser ask [--source-cli /absolute/path/to/dist/cli.js] [--cwd /absolute/path/to/repo] [--target-url url --confirm-target] [--new-chat] [--stdin] [--json] [--auto-login|--no-auto-login] [--file path] [--attach path] [--tool deep-research|web-search|create-image] [--model Pro] [--pro-mode 기본|확장] [--effort 즉시|중간|높음|"매우 높음"] [--project "name" | --project-new "name"] "prompt"
   prodex pro latest [--source-cli /absolute/path/to/dist/cli.js] [--cwd /absolute/path/to/repo]
   prodex pro list [--source-cli /absolute/path/to/dist/cli.js] [--cwd /absolute/path/to/repo] [--json]
   prodex pro show <task-id|latest> [--source-cli /absolute/path/to/dist/cli.js] [--cwd /absolute/path/to/repo]
@@ -203,7 +203,7 @@ export function printTasksHelp(stdout) {
     stdout(`prodex tasks
 
 Commands:
-  prodex tasks create [--cwd /absolute/path/to/repo] --title "Title" --prompt "Prompt" [--repo-id id] [--file path]
+  prodex tasks create [--cwd /absolute/path/to/repo] --title "Title" --prompt "Prompt" [--repo-id id] [--file path] [--attach path] [--tool deep-research|web-search|create-image]
   prodex tasks list [--status new|claimed|done|blocked] [--cwd /absolute/path/to/repo] [--json]
   prodex tasks show <task-id|latest> [--cwd /absolute/path/to/repo]
   prodex tasks claim <task-id> [--cwd /absolute/path/to/repo] [--by codex]
@@ -252,8 +252,8 @@ export function printProBrowserHelp(stdout, sourceCli) {
         : "prodex pro browser smoke [--source-cli /absolute/path/to/dist/cli.js] [--cwd /absolute/path/to/repo] [--port 9333] [--timeout-ms 90000]";
     const selectionUsage = '[--model Pro] [--pro-mode 기본|확장] [--effort 즉시|중간|높음|"매우 높음"] [--project "name" | --project-new "name"]';
     const askUsage = sourceCli
-        ? `${cli} pro browser ask${sourceCliOption} [--cwd /absolute/path/to/repo] [--port 9333] [--timeout-ms 300000] [--busy-wait-ms 600000] [--target-url url --confirm-target] [--new-chat] [--stdin] [--json] [--auto-login|--no-auto-login] [--file path] ${selectionUsage} "prompt"`
-        : `prodex pro browser ask [--source-cli /absolute/path/to/dist/cli.js] [--cwd /absolute/path/to/repo] [--port 9333] [--timeout-ms 300000] [--busy-wait-ms 600000] [--target-url url --confirm-target] [--new-chat] [--stdin] [--json] [--auto-login|--no-auto-login] [--file path] ${selectionUsage} "prompt"`;
+        ? `${cli} pro browser ask${sourceCliOption} [--cwd /absolute/path/to/repo] [--port 9333] [--timeout-ms 300000] [--busy-wait-ms 600000] [--target-url url --confirm-target] [--new-chat] [--stdin] [--json] [--auto-login|--no-auto-login] [--file path] [--attach path] [--tool deep-research|web-search|create-image] ${selectionUsage} "prompt"`
+        : `prodex pro browser ask [--source-cli /absolute/path/to/dist/cli.js] [--cwd /absolute/path/to/repo] [--port 9333] [--timeout-ms 300000] [--busy-wait-ms 600000] [--target-url url --confirm-target] [--new-chat] [--stdin] [--json] [--auto-login|--no-auto-login] [--file path] [--attach path] [--tool deep-research|web-search|create-image] ${selectionUsage} "prompt"`;
     const modelsUsage = sourceCli
         ? `${cli} pro browser models${sourceCliOption} [--port 9333] [--timeout-ms 15000]`
         : "prodex pro browser models [--source-cli /absolute/path/to/dist/cli.js] [--port 9333] [--timeout-ms 15000]";

package/dist/cli-pro.js

@@ -1,7 +1,8 @@
+import { existsSync, statSync } from "node:fs";
 import { mkdir, readFile, writeFile } from "node:fs/promises";
 import path from "node:path";
 import { buildDryRunBundle } from "./bundle.js";
-import { DEFAULT_CDP_PORT, resolveCdpPort, chatGptVisibilityBlocker, defaultChatGptProfileDir, formatDurationMs, getChatGptBrowserStatus, listChatGptModelOptions, listChatGptSidebarProjects, normalizeChatGptTargetUrl, openChatGptBrowser, parseProMode, parseReasoningEffort, ensureVirtualDisplay, minimizeChatGptWindow, readLastBrowserLoginLaunch, resolveVirtualDisplayPreference, resolveHeadlessPreference, recordBrowserLoginLaunch, recoverChatGptAnswerFromThread, sendChatGptPrompt } from "./chatgpt-browser.js";
+import { DEFAULT_CDP_PORT, resolveCdpPort, chatGptVisibilityBlocker, defaultChatGptProfileDir, formatDurationMs, getChatGptBrowserStatus, listChatGptModelOptions, listChatGptSidebarProjects, normalizeChatGptTargetUrl, openChatGptBrowser, parseProMode, parseReasoningEffort, defaultTimeoutForTools, ensureVirtualDisplay, minimizeChatGptWindow, readLastBrowserLoginLaunch, resolveVirtualDisplayPreference, resolveHeadlessPreference, recordBrowserLoginLaunch, recoverChatGptAnswerFromThread, sendChatGptPrompt } from "./chatgpt-browser.js";
 import { ASK_PRO_BOOLEAN_FLAGS, ASK_PRO_PREVIEW_VALUE_FLAGS, ASK_PRO_VALUE_FLAGS, assertHelpRequestArgs, assertNoExtraArgs, assertOnlyOptions, findHelpFlagIndexBeforePromptDelimiter, formatCliCommand, hasAskProDryRunMode, hasAskProMode, hasAskProSendMode, isHelpSubcommand, parseAskProArgs, printHelpIfRequested, readFlag, readPortFlag, readPositionalsWithOptions, readNonNegativeIntegerFlag, readPositiveIntegerFlag, readRepeatedFlag, resolveCwdFlag, resolveOptionalFileFlag, unknownSubcommandError } from "./cli-args.js";
 import { printProBrowserHelp, printProHelp } from "./cli-help.js";
 import { listRawResultsForInspection, listTasksForInspection } from "./cli-ledger.js";
@@ -722,6 +723,29 @@ export async function runAskProCommand(rest, io) {
         }
         return rel;
     });
+    // --attach UPLOADS the file (pdf, pptx, image) instead of inlining its
+    // text like --file. Same escape guard: an agent must not be able to upload
+    // ~/.ssh or anything else outside the repo to a chat.
+    const attachments = readRepeatedFlag(parsedAskPro.optionArgs, "--attach").map((file) => {
+        const absolute = path.resolve(targetCwd, file);
+        const rel = path.relative(targetCwd, absolute);
+        if (rel === "" || rel.startsWith("..") || path.isAbsolute(rel)) {
+            throw new Error(`--attach "${file}" is outside the repo root (${targetCwd}). Pass a path inside the repo, or point --cwd at that repo.`);
+        }
+        if (!existsSync(absolute) || !statSync(absolute).isFile()) {
+            throw new Error(`--attach "${file}" is not a readable file (looked at ${absolute}).`);
+        }
+        return absolute;
+    });
+    if (attachments.length > 0 && !hasSendMode) {
+        throw new Error("--attach only applies when sending (`prodex pro browser ask`); the dry-run preview cannot upload files.");
+    }
+    // --tool turns on a composer tool (deep-research, web-search,
+    // create-image, or any label the menu shows) for this send.
+    const tools = readRepeatedFlag(parsedAskPro.optionArgs, "--tool");
+    if (tools.length > 0 && !hasSendMode) {
+        throw new Error("--tool only applies when sending (`prodex pro browser ask`); the dry-run preview cannot open ChatGPT's tools menu.");
+    }
     const targetUrl = readFlag(parsedAskPro.optionArgs, "--target-url");
     const normalizedTargetUrl = targetUrl ? normalizeChatGptTargetUrl(targetUrl) : undefined;
     if (!normalizedTargetUrl && parsedAskPro.optionArgs.includes("--confirm-target")) {
@@ -815,7 +839,7 @@ export async function runAskProCommand(rest, io) {
     // consults (observed in several field sessions).
     const defaultBrowserTimeoutMs = effectiveProSelection ? 1_200_000 : 300_000;
     const browserTimeoutMs = hasSendMode
-        ? (readPositiveIntegerFlag(parsedAskPro.optionArgs, "--timeout-ms") ?? defaultBrowserTimeoutMs)
+        ? (readPositiveIntegerFlag(parsedAskPro.optionArgs, "--timeout-ms") ?? defaultTimeoutForTools(tools, defaultBrowserTimeoutMs))
         : undefined;
     const sourceCli = resolveOptionalFileFlag(io.cwd, parsedAskPro.optionArgs, "--source-cli");
     const bundle = await buildDryRunBundle(targetCwd, { prompt: promptText, files });
@@ -880,6 +904,8 @@ export async function runAskProCommand(rest, io) {
             prompt: bundle.text,
             targetUrl: normalizedTargetUrl,
             timeoutMs: browserTimeoutMs,
+            ...(attachments.length > 0 ? { attachments } : {}),
+            ...(tools.length > 0 ? { tools } : {}),
             ...(newChat ? { newChat: true } : {}),
             ...(busyWaitMs !== undefined ? { busyWaitMs } : {}),
             project: selectionProject,
@@ -1148,6 +1174,8 @@ export async function performBrowserConsultForMcp(cwd, input, onProgress) {
         ...(input.project !== undefined ? ["--project", input.project] : []),
         ...(input.timeout_ms !== undefined ? ["--timeout-ms", String(input.timeout_ms)] : []),
         ...(input.files ?? []).flatMap((file) => ["--file", file]),
+        ...(input.attach ?? []).flatMap((file) => ["--attach", file]),
+        ...(input.tools ?? []).flatMap((tool) => ["--tool", tool]),
         ...(input.new_chat ? ["--new-chat"] : []),
         "--",
         input.prompt

package/dist/mcp.js

@@ -150,6 +150,16 @@ export function createServer(cwd = process.cwd(), options = {}) {
                 project: McpShortTextSchema.optional(),
                 timeout_ms: z.number().int().positive().max(3_600_000).optional(),
                 files: z.array(McpShortTextSchema).max(20).optional(),
+                tools: z
+                    .array(McpShortTextSchema)
+                    .max(4)
+                    .optional()
+                    .describe("ChatGPT composer tools to enable for this consult: \"deep-research\" (a multi-minute browsed report - the timeout rises to 30 minutes automatically), \"web-search\" (current facts), \"create-image\". Deep research often replies with a CLARIFYING QUESTION first; answer it with a normal follow-up consult in the same thread."),
+                attach: z
+                    .array(McpShortTextSchema)
+                    .max(10)
+                    .optional()
+                    .describe("Repo-relative paths to UPLOAD to ChatGPT as real attachments (pdf, pptx, xlsx, images, or any file you want ChatGPT to parse itself). Use this instead of `files` for binaries and for large documents; `files` inlines a text file's contents into the prompt, which cannot carry a binary and bloats the prompt."),
                 new_chat: z
                     .boolean()
                     .optional()

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@youdie006/prodex",
-  "version": "0.19.2",
+  "version": "0.21.0",
   "description": "Local receipt bus for coordinating Codex execution with ChatGPT Pro/Projects consultation.",
   "author": "youdie006",
   "license": "MIT",