@ferris1225/pi-subagents 2.2.0 → 2.3.0

package/README.md

@@ -239,6 +239,15 @@ currently unavailable vision selection is skipped immediately without rewriting
 it. A vision-flagged auto-fix chain keeps the flag for worker/re-review rounds
 because they may need to inspect the same images.
 
+As a deterministic fallback, a brief that names an image file (an extension such
+as `.png`, `.jpg`, `.webp`) or describes frontend/UI work — Vue/React/Svelte
+pages, mockups, screenshots, page styling — is dispatched as
+vision even without the flag: the model cannot be swapped mid-run, so a worker
+that screenshots and visually verifies what it builds must start on a
+vision-capable model. How do you verify which model ran? The live widget line,
+dispatch result row, and `subagent_status` all show each run's effective model
+id, and a selected→main handoff is labeled with its origin.
+
 ### Controlling and stopping
 
 Dispatch confirmations, tool result rows, and completion blocks all show the

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@ferris1225/pi-subagents",
-  "version": "2.2.0",
+  "version": "2.3.0",
   "description": "Controllable background sub-agent threads for pi: specialized roles, capability-aware thinking, direct main-model fallback, auto-fix chains, and Git worktree isolation.",
   "type": "module",
   "license": "MIT",

package/src/dispatch.ts

@@ -59,7 +59,7 @@ export { FORK_CONTINUATION_PROMPT, isWorktreeCapableAgent } from "./thread-lifec
 const NON_BLANK_TASK_OPTIONS = { minLength: 1, pattern: "\\S" } as const;
 
 const VISION_DESCRIPTION =
-	"Set true when the task may require viewing images (screenshots, mockups, designs) — the configured vision model is used first, then model-level failures hand directly to the current main-window model";
+	"Set true when the task may require viewing images (screenshots, mockups, designs) — the configured vision model is used first, then model-level failures hand directly to the current main-window model. A task that names an image file or describes frontend/UI work (Vue/React, mockups, screenshots, page styling) is dispatched as vision automatically";
 
 const ISOLATION_DESCRIPTION =
 	"Filesystem isolation: shared uses the caller's working tree; worktree creates a detached temporary Git worktree (write-capable agents, including worker and cleaner, only)";
@@ -95,6 +95,24 @@ export function defaultIsolationMode(mode: "single" | "parallel", agentName: str
 	return mode === "parallel" && agentName === "worker" ? "worktree" : "shared";
 }
 
+const IMAGE_FILE_PATTERN = /\.(?:png|jpe?g|webp|gif|bmp|tiff?|avif)\b/i;
+/** Frontend/UI work: the agent may screenshot and visually verify what it
+ * builds, and the model cannot be swapped mid-run — vision must be on at
+ * spawn. `\bUI\b` stays case-sensitive so ordinary words never match. */
+const FRONTEND_PATTERN = /\b(?:vue|react|svelte|angular|nuxt|next\.?js|vite|tailwind|screenshots?|mockups?|wireframes?|UI)\b|前端|界面|页面|样式|截图|设计稿/i;
+const FRONTEND_FILE_PATTERN = /\.(?:vue|html?|css|svg)\b/i;
+
+/** Deterministic vision fallback: a brief that names an image file or describes
+ * frontend/UI work is vision work even when the dispatcher forgot the flag —
+ * prompting alone cannot guarantee `vision: true`, and a non-vision model can
+ * neither see reference images nor visually verify what it builds. */
+export function taskImpliesVision(task: string): boolean {
+	return IMAGE_FILE_PATTERN.test(task) || FRONTEND_FILE_PATTERN.test(task) || FRONTEND_PATTERN.test(task);
+}
+
+const wantsVision = (flag: boolean | undefined, task: string): boolean =>
+	flag === true || taskImpliesVision(task);
+
 const autoFixRootTails = new Map<string, Promise<void>>();
 
 async function canonicalAutoFixRoot(cwd: string): Promise<string> {
@@ -628,7 +646,7 @@ export function registerSubagentTool(pi: ExtensionAPI, runtime: SubagentRuntime)
 						item.agent,
 						item.task,
 						item.cwd,
-						item.vision === true,
+						wantsVision(item.vision, item.task),
 						defaultIsolationMode("parallel", item.agent, item.isolation as IsolationMode | undefined),
 					));
 				}
@@ -666,7 +684,7 @@ export function registerSubagentTool(pi: ExtensionAPI, runtime: SubagentRuntime)
 				params.agent as string,
 				params.task as string,
 				params.cwd,
-				params.vision === true,
+				wantsVision(params.vision, params.task as string),
 				defaultIsolationMode("single", params.agent as string, params.isolation as IsolationMode | undefined),
 			);
 			if (result.exitCode !== -1) {

package/src/prompt.ts

@@ -67,7 +67,7 @@ ${hasMultiple ? `- Run INDEPENDENT tasks in parallel: one subagent call with a \
 - Trust but verify: a sub-agent's summary describes intent, not outcome. Check the actual changes/results before reporting work done.
 ${hasExplore ? "- Treat `explore` findings as a retrieval index, never as sole proof for edits, deletion, security, compatibility, persistence, or dynamic reachability. Re-read load-bearing files before acting. An underpowered model can be false economy on complex dynamic, concurrent, migration, or security-sensitive code; use a stronger model or specialist there.\n" : ""}
 Vision tasks:
-- Judge whether a delegated task may require viewing images (frontend screenshots, mockups, design files, visual regression comparisons). If it might, pass \`vision: true\` in the subagent call and give the sub-agent the exact image paths — it reads them with its read tool.
+- Judge whether a delegated task may require viewing images (frontend screenshots, mockups, design files, visual regression comparisons). If it might, pass \`vision: true\` in the subagent call and give the sub-agent the exact image paths — it reads them with its read tool. Naming an image file or describing frontend/UI work (Vue/React page, mockup, page styling) dispatches as vision automatically, but keep passing the flag for image work described without those signals.
 - \`vision: true\` runs the sub-agent on the vision-capable model configured in /subagents-setup; when none is configured it falls back to the main session's current model. Do not skip the flag because the agent's default model looks fast — a non-vision model cannot see the images.
 
 Result handoff (do not re-state):