@lazyingart/agintiflow 0.8.8 → 0.8.9

package/README.md

@@ -50,6 +50,7 @@ On first interactive use, if no DeepSeek key is detected, `aginti` asks you to p
 
 ```bash
 aginti login deepseek
+# inside chat, use /login or /auth
 # or non-interactively:
 printf '%s' "$DEEPSEEK_API_KEY" | aginti keys set deepseek --stdin
 ```
@@ -62,7 +63,7 @@ aginti
 aginti chat
 ```
 
-Inside chat, type normal requests such as `write a small Python CLI app with tests`. The default is Docker workspace mode with approved package installs, so coding, plotting, and LaTeX tasks can set up project-local tools without touching the host. Use `/help` for commands, `/latex on` for PDF work, `/docker off` only when you intentionally want host mode, `/sessions` to list project runs, and `/resume latest` or `/resume <session-id>` to continue work. Esc or Ctrl+C stops the active run cleanly and prints the resume command.
+Inside chat, type normal requests such as `write a small Python CLI app with tests`. The default is Docker workspace mode with approved package installs, so coding, plotting, and LaTeX tasks can set up project-local tools without touching the host. Use `/help` for commands, `/login` or `/auth` to paste a provider key, `/latex on` for PDF work, `/docker off` only when you intentionally want host mode, `/sessions` to list project runs, and `/resume latest` or `/resume <session-id>` to continue work. Type `/` then Tab for command completion. Esc or Ctrl+C stops the active run cleanly and prints the resume command.
 
 Launch the local web UI from an installed package:
 

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@lazyingart/agintiflow",
-  "version": "0.8.8",
+  "version": "0.8.9",
   "type": "module",
   "description": "AgInTiFlow is a resumable Playwright website-control agent with OpenAI-compatible tool calling.",
   "license": "Apache-2.0",

package/src/agent-runner.js

@@ -258,6 +258,7 @@ function createInitialState(config, sessionId) {
         content: [
           "You are a careful browser and shell agent with a small tool surface.",
           "Use only the provided tools.",
+          "The execution plan is not the final answer. After planning, actively use tools until the requested task is complete or genuinely blocked.",
           "If the shell tool can satisfy a local task, prefer it before opening a browser.",
           "Do not open a browser page just because a start URL exists. Treat it as a suggestion only.",
           "Only reference element ids from the latest browser snapshot.",
@@ -279,6 +280,9 @@ function createInitialState(config, sessionId) {
           "A frontend canvas/artifacts tunnel exists. Use send_to_canvas when important markdown, diffs, screenshots, images, or workspace files should be highlighted in the UI. It is optional and ordinary final text can still go directly to finish.",
           "For visual-output requests such as draw, plot, graph, chart, diagram, figure, image, or visualization, proactively publish a canvas artifact even when the user does not mention canvas. If workspace file tools are enabled, prefer creating a small SVG or markdown artifact and call send_to_canvas with selected=true.",
           "Work like a practical coding agent: inspect when useful, edit with file tools, run safe checks when they add confidence, and keep outputs inside the workspace.",
+          "For large projects, decompose into useful files and milestones, implement a coherent minimal version first, then iterate with checks rather than only describing what you would do.",
+          "For website/app/code/LaTeX/Python/C/shell tasks, create or edit real workspace files, run available build/compile/test commands, and surface artifacts through the canvas when useful.",
+          "For research or web-search tasks, use browser tools or safe shell network tools when the current policy allows; cite or save useful sources in workspace notes when the task needs traceability.",
           "Use the canvas tunnel for outputs the user would likely want to inspect visually, such as figures, PDFs, screenshots, images, important markdown, or generated files.",
           "For environment or system-maintenance work, use the configured sandbox and package policy; Docker workspace mode is the preferred place for installs and toolchain setup.",
           "If the user asks to open a generated local website or file, use open_workspace_file for a file or preview_workspace for a static site. Do not keep retrying the same localhost URL when a preview fails.",
@@ -309,6 +313,7 @@ function createInitialState(config, sessionId) {
           "Canvas/artifacts tunnel: available through send_to_canvas for optional frontend rendering.",
           "Visual-output requests should produce a canvas artifact without requiring the user to ask for canvas explicitly.",
           "Use file, shell, browser, canvas, and wrapper tools when they are useful; choose the workflow from the user's request.",
+          "Do not stop at a plan when tools can accomplish the request. Continue through implementation, checks, artifact selection, and finish.",
           "Use the configured sandbox and package policy for environment or system-maintenance work.",
         ]
           .filter(Boolean)
@@ -1078,6 +1083,7 @@ export async function runAgent(config) {
         url: snapshot.url,
         title: snapshot.title,
         screenshotPath: snapshot.screenshotPath,
+        screenshotWarning: snapshot.screenshotWarning || "",
         snapshotPath: snapshot.snapshotPath,
       });
       observers.event("snapshot.captured", {
@@ -1085,6 +1091,7 @@ export async function runAgent(config) {
         url: snapshot.url,
         title: snapshot.title,
         screenshotPath: snapshot.screenshotPath,
+        screenshotWarning: snapshot.screenshotWarning || "",
       });
 
       state.messages.push({

package/src/interactive-cli.js

@@ -3,10 +3,10 @@ import { emitKeypressEvents } from "node:readline";
 import { stdin as input, stdout as output } from "node:process";
 import { runAgent } from "./agent-runner.js";
 import { loadConfig } from "./config.js";
-import { initProject, listProjectSessions } from "./project.js";
+import { initProject, listProjectSessions, providerKeyStatus, setProviderKey } from "./project.js";
 import { normalizePackageInstallPolicy, normalizeSandboxMode } from "./command-policy.js";
 import { normalizeTaskProfile } from "./task-profiles.js";
-import { promptAndSaveDeepSeekKey, shouldPromptForDeepSeek } from "./auth-onboarding.js";
+import { promptAndSaveDeepSeekKey, promptHidden, shouldPromptForDeepSeek } from "./auth-onboarding.js";
 
 const useColor = Boolean(input.isTTY && output.isTTY && process.env.AGINTIFLOW_NO_COLOR !== "1");
 const ansi = {
@@ -25,6 +25,26 @@ const ansi = {
   systemBg: "\x1b[48;5;236m\x1b[38;5;245m",
 };
 const brandPalette = ["\x1b[38;5;45m", "\x1b[38;5;81m", "\x1b[38;5;86m", "\x1b[38;5;118m", "\x1b[38;5;226m"];
+const SLASH_COMMANDS = [
+  "/help",
+  "/status",
+  "/login",
+  "/auth",
+  "/new",
+  "/resume",
+  "/sessions",
+  "/profile",
+  "/routing",
+  "/provider",
+  "/model",
+  "/docker",
+  "/latex",
+  "/installs",
+  "/cwd",
+  "/init",
+  "/web",
+  "/exit",
+];
 
 function color(value, ...codes) {
   if (!useColor || codes.length === 0) return String(value);
@@ -43,6 +63,13 @@ function userPrompt() {
   return `\n${label("user>", ansi.userBg)} `;
 }
 
+function commandCompleter(line = "") {
+  const trimmed = String(line || "");
+  if (!trimmed.startsWith("/")) return [[], trimmed];
+  const hits = SLASH_COMMANDS.filter((command) => command.startsWith(trimmed));
+  return [hits.length > 0 ? hits : SLASH_COMMANDS, trimmed];
+}
+
 function stripMarkdown(text) {
   const lines = String(text || "").split(/\r?\n/);
   let inFence = false;
@@ -147,6 +174,8 @@ function printHelp() {
       "Commands:",
       "  /help                     Show this help.",
       "  /status                   Show active route, workspace, sandbox, and session.",
+      "  /login [deepseek|openai]  Paste and save a project-local API key.",
+      "  /auth [deepseek|openai]   Alias for /login.",
       "  /new                      Start a fresh session on the next message.",
       "  /resume <session-id>      Continue a saved session.",
       "  /sessions                 List recent sessions in this project.",
@@ -162,6 +191,7 @@ function printHelp() {
       "  /exit                     Quit.",
       "",
       "Type a normal request to run the agent. Example: write a Python CLI app with tests",
+      "Type / then Tab to autocomplete commands.",
       "While a run is active, press Esc or Ctrl+C once to stop gracefully and print a resume command.",
     ].join("\n")
   );
@@ -292,6 +322,28 @@ async function maybeOnboardDeepSeekKey(state) {
   printAgentMessage("No key saved. Continuing in local mock mode. Use `/provider deepseek` after running `aginti login deepseek`.");
 }
 
+async function promptAndSaveProviderKey(provider = "deepseek", state = null) {
+  const normalized = ["openai", "deepseek"].includes(String(provider || "").toLowerCase())
+    ? String(provider || "").toLowerCase()
+    : "deepseek";
+  const labelText = normalized === "openai" ? "OpenAI" : "DeepSeek";
+  const key = await promptHidden(`${labelText} API key/token (paste, Enter to save): `);
+  if (!key) {
+    printAgentMessage("No key saved.");
+    return;
+  }
+
+  const result = await setProviderKey(process.cwd(), normalized, key);
+  if (state) {
+    state.provider = normalized;
+    if (state.routingMode === "manual" && state.model === "mock-agent") {
+      state.routingMode = "smart";
+      state.model = "";
+    }
+  }
+  printAgentMessage(`Saved ${result.keyName} to project-local ignored env. Raw key was not printed.`);
+}
+
 async function handleCommand(line, state, packageDir) {
   const [command, ...rest] = line.slice(1).trim().split(/\s+/);
   const value = rest.join(" ").trim();
@@ -303,6 +355,12 @@ async function handleCommand(line, state, packageDir) {
   if (command === "exit" || command === "quit" || command === "q") return false;
   if (command === "status") {
     printStatus(state);
+    const keys = providerKeyStatus(process.cwd());
+    printSystemLine(`keys deepseek=${keys.deepseek ? "available" : "missing"} openai=${keys.openai ? "available" : "missing"}`);
+    return true;
+  }
+  if (command === "login" || command === "auth") {
+    await promptAndSaveProviderKey(value || "deepseek", state);
     return true;
   }
   if (command === "new") {
@@ -410,7 +468,13 @@ async function handleCommand(line, state, packageDir) {
     return true;
   }
 
-  printAgentMessage(`Unknown command: /${command}. Use /help.`);
+  const typed = `/${command}`;
+  const suggestions = SLASH_COMMANDS.filter((candidate) => candidate.startsWith(typed));
+  printAgentMessage(
+    suggestions.length > 0
+      ? `Unknown command: /${command}. Did you mean:\n${suggestions.map((item) => `  ${item}`).join("\n")}`
+      : `Unknown command: /${command}. Use /help.`
+  );
   return true;
 }
 
@@ -500,7 +564,12 @@ async function runPrompt(prompt, state, packageDir) {
 
 export async function startInteractiveCli(args = {}, { packageDir, packageVersion } = {}) {
   const state = createState(args);
-  const rl = readline.createInterface({ input, output, terminal: Boolean(input.isTTY && output.isTTY) });
+  const rl = readline.createInterface({
+    input,
+    output,
+    terminal: Boolean(input.isTTY && output.isTTY),
+    completer: commandCompleter,
+  });
 
   await renderLaunchHeader(packageVersion);
   printSystemLine(`Project: ${process.cwd()}`);

package/src/model-client.js

@@ -154,7 +154,7 @@ export async function createPlan(client, config, state) {
       {
         role: "system",
         content:
-          "You are planning a browser-and-shell automation task. If a local shell command can satisfy the goal, prefer that before browser actions. Treat any suggested start URL as optional. Write a concise execution plan with 3 to 6 steps. Mention risks or blockers when relevant. Keep it short and practical.",
+          "You are planning a browser, shell, workspace, and coding-agent task. The plan is only a launchpad: after planning, the runtime will continue with tools until the task is complete or genuinely blocked. Prefer real workspace edits/checks over advice-only answers. If a local shell command can satisfy the goal, prefer that before browser actions. Treat any suggested start URL as optional. Write a concise execution plan with 3 to 6 steps. Mention risks or blockers when relevant. Keep it short and practical.",
       },
       {
         role: "user",
@@ -174,6 +174,8 @@ export async function createPlan(client, config, state) {
           `Task profile: ${taskProfile.label}. ${taskProfile.prompt}`,
           "A canvas/artifacts tunnel is available through send_to_canvas. Use it when an output should be highlighted visually, such as screenshots, image files, important markdown, diffs, or generated artifact paths. It is optional for ordinary text answers.",
           "Work like a practical coding agent: inspect when useful, edit with file tools, run safe checks when they add confidence, and keep outputs inside the workspace.",
+          "For large apps, websites, LaTeX documents, Python/C/shell projects, or system tasks, plan a coherent minimal implementation, then use tools to create files, run checks, and publish artifacts.",
+          "For web search or current information tasks, plan to use browser tools or safe shell network tools when allowed, then preserve useful source notes if the output depends on them.",
           "Use the canvas tunnel for outputs the user would likely want to inspect visually, such as figures, PDFs, screenshots, images, important markdown, or generated files.",
           "For environment or system-maintenance work, prefer project-local dry-run plans/scripts unless the configured policy explicitly allows stronger actions.",
           "Docker language/toolchain installs should prefer /aginti-env or project files so they persist across runs; apt/apk changes are ephemeral unless the image is rebuilt.",

package/src/snapshot.js

@@ -1,55 +1,71 @@
 export async function captureSnapshot(page, store, step) {
-  const snapshot = await page.evaluate(() => {
-    const isVisible = (el) => {
-      const rect = el.getBoundingClientRect();
-      const style = window.getComputedStyle(el);
-      return rect.width > 0 && rect.height > 0 && style.display !== "none" && style.visibility !== "hidden";
-    };
-
-    document.querySelectorAll("[data-agent-id]").forEach((el) => el.removeAttribute("data-agent-id"));
-
-    const candidates = Array.from(
-      document.querySelectorAll("a, button, input, textarea, select, [role='button'], [contenteditable='true']")
-    );
-
-    const elements = [];
-
-    for (const el of candidates) {
-      if (!isVisible(el)) continue;
-      if (elements.length >= 50) break;
-
-      const id = String(elements.length + 1);
-      el.setAttribute("data-agent-id", id);
-
-      const tag = el.tagName.toLowerCase();
-      elements.push({
-        id,
-        tag,
-        role: el.getAttribute("role") || "",
-        text: (el.innerText || el.textContent || "").replace(/\s+/g, " ").trim().slice(0, 100),
-        ariaLabel: (el.getAttribute("aria-label") || "").trim().slice(0, 100),
-        placeholder: (el.getAttribute("placeholder") || "").trim().slice(0, 100),
-        href: tag === "a" ? (el.getAttribute("href") || "").trim() : "",
-        inputType: tag === "input" ? (el.getAttribute("type") || "text").trim() : "",
-        autocomplete: (el.getAttribute("autocomplete") || "").trim(),
-      });
-    }
-
-    return {
-      title: document.title,
-      url: window.location.href,
-      pageText: (document.body?.innerText || "").replace(/\s+/g, " ").trim().slice(0, 2500),
-      elements,
-    };
-  });
+  const snapshot = await page
+    .evaluate(() => {
+      const isVisible = (el) => {
+        const rect = el.getBoundingClientRect();
+        const style = window.getComputedStyle(el);
+        return rect.width > 0 && rect.height > 0 && style.display !== "none" && style.visibility !== "hidden";
+      };
+
+      document.querySelectorAll("[data-agent-id]").forEach((el) => el.removeAttribute("data-agent-id"));
+
+      const candidates = Array.from(
+        document.querySelectorAll("a, button, input, textarea, select, [role='button'], [contenteditable='true']")
+      );
+
+      const elements = [];
+
+      for (const el of candidates) {
+        if (!isVisible(el)) continue;
+        if (elements.length >= 50) break;
+
+        const id = String(elements.length + 1);
+        el.setAttribute("data-agent-id", id);
+
+        const tag = el.tagName.toLowerCase();
+        elements.push({
+          id,
+          tag,
+          role: el.getAttribute("role") || "",
+          text: (el.innerText || el.textContent || "").replace(/\s+/g, " ").trim().slice(0, 100),
+          ariaLabel: (el.getAttribute("aria-label") || "").trim().slice(0, 100),
+          placeholder: (el.getAttribute("placeholder") || "").trim().slice(0, 100),
+          href: tag === "a" ? (el.getAttribute("href") || "").trim() : "",
+          inputType: tag === "input" ? (el.getAttribute("type") || "text").trim() : "",
+          autocomplete: (el.getAttribute("autocomplete") || "").trim(),
+        });
+      }
+
+      return {
+        title: document.title,
+        url: window.location.href,
+        pageText: (document.body?.innerText || "").replace(/\s+/g, " ").trim().slice(0, 2500),
+        elements,
+      };
+    })
+    .catch((error) => ({
+      title: "Browser page open",
+      url: page.url(),
+      pageText: `DOM snapshot unavailable: ${error instanceof Error ? error.message : String(error)}`.slice(0, 2500),
+      elements: [],
+      snapshotWarning: "dom-unavailable",
+    }));
 
   const screenshotPath = store.screenshotPath(step);
-  await page.screenshot({ path: screenshotPath, fullPage: true });
-  const snapshotPath = await store.saveSnapshot(step, snapshot);
+  let savedScreenshotPath = screenshotPath;
+  let screenshotWarning = "";
+  try {
+    await page.screenshot({ path: screenshotPath, fullPage: true });
+  } catch (error) {
+    savedScreenshotPath = "";
+    screenshotWarning = `Screenshot unavailable: ${error instanceof Error ? error.message : String(error)}`.slice(0, 500);
+  }
+  const snapshotPath = await store.saveSnapshot(step, { ...snapshot, screenshotWarning });
 
   return {
     ...snapshot,
-    screenshotPath,
+    screenshotPath: savedScreenshotPath,
+    screenshotWarning,
     snapshotPath,
   };
 }

package/src/task-profiles.js

@@ -3,14 +3,14 @@ export const TASK_PROFILES = {
     id: "auto",
     label: "Auto",
     prompt:
-      "Infer the task type from the user request. Prefer the smallest safe tool sequence, preserve workspace files, and summarize what changed.",
+      "Infer the task type from the user request. Prefer the smallest safe tool sequence that actually completes the work, preserve workspace files, run useful checks, and summarize what changed.",
     tools: ["browser", "shell", "files", "canvas"],
   },
   code: {
     id: "code",
     label: "Code writing",
     prompt:
-      "Act like a coding agent: understand the request, edit workspace files, run useful safe checks, and report changed files and residual risks.",
+      "Act like a coding agent: understand the request, edit workspace files, run useful safe checks, iterate on failures, and report changed files and residual risks.",
     tools: ["files", "shell", "sandbox"],
   },
   writing: {
@@ -52,7 +52,7 @@ export const TASK_PROFILES = {
     id: "website",
     label: "Website testing",
     prompt:
-      "For website-testing tasks, create or inspect the site, add a local check when useful, and use the configured sandbox/package policy for dependencies.",
+      "For website/app tasks, create or inspect real site files, preview with workspace preview tools, add local checks when useful, and use the configured sandbox/package policy for dependencies.",
     tools: ["files", "shell", "canvas", "sandbox"],
   },
   aaps: {
@@ -66,7 +66,7 @@ export const TASK_PROFILES = {
     id: "latex",
     label: "LaTeX",
     prompt:
-      "For LaTeX/PDF tasks, create source and figures in a subfolder, compile when a TeX toolchain is available, and send the PDF through the canvas tunnel. In Docker, use /workspace for project outputs and the persistent Python/conda/tool cache under /aginti-env when setup is needed.",
+      "For LaTeX/PDF tasks, locate or create source and figures in a subfolder, compile when a TeX toolchain is available, run enough passes for references, and send the PDF through the canvas tunnel. In Docker, use /workspace for project outputs and the persistent Python/conda/tool cache under /aginti-env when setup is needed.",
     tools: ["files", "shell", "canvas", "sandbox"],
   },
   maintenance: {