What causes "${formatTimeoutMessage()}" in can1357/oh-my-pi?

Increase inspect_image.timeoutMs in settings (e.g. 60000 or more).

can1357/oh-my-pi · error · ToolError

${formatTimeoutMessage()}

Error message

${formatTimeoutMessage()}

What it means

The one-shot vision LLM request aborted with AbortError/TimeoutError and the tool's own timeout signal (inspect_image.timeoutMs) was the cause, so it converts the abort into a descriptive ToolError stating the elapsed time and how to raise or disable the timeout. If the caller's signal aborted instead, the original error is rethrown.

Source

Thrown at packages/coding-agent/src/tools/inspect-image.ts:311

							role: "user",
							content: [
								{ type: "image", data: imageInput.data, mimeType: imageInput.mimeType },
								{ type: "text", text: params.question },
							],
							timestamp: Date.now(),
						},
					],
				},
				{
					apiKey: modelRegistry.resolver(model, this.session.getSessionId?.() ?? undefined),
					signal: effectiveSignal,
					reasoning,
				},
				{ telemetry, oneshotKind: "inspect_image", completeImpl: this.completeImageRequest },
			);
		} catch (error) {
			if (error instanceof Error && (error.name === "AbortError" || error.name === "TimeoutError")) {
				if (timedOut()) throw new ToolError(formatTimeoutMessage());
			}
			throw error;
		}

		if (response.stopReason === "error") {
			throw new ToolError(response.errorMessage ?? "inspect_image request failed.");
		}
		if (response.stopReason === "aborted") {
			if (timedOut()) throw new ToolError(formatTimeoutMessage());
			throw new ToolError("inspect_image request aborted.");
		}

		const text = extractTextContent(response);
		if (!text) {
			throw new ToolError("inspect_image model returned no text output.");
		}

		return {

View on GitHub (pinned to 9690622007)

Solutions

Increase inspect_image.timeoutMs in settings (e.g. 60000 or more).
Set inspect_image.timeoutMs to 0 to disable the timeout entirely.
Check provider status/latency or switch to a faster vision model via modelRoles.vision.
Reduce image size to cut upload and inference time.

Example fix

// before
{ "inspect_image": { "timeoutMs": 5000 } }
// after
{ "inspect_image": { "timeoutMs": 120000 } }

Defensive patterns

Strategy: retry

Validate before calling

const timeoutMs = session.settings.get("inspect_image.timeoutMs");
if (typeof timeoutMs === "number" && timeoutMs > 0 && timeoutMs < 30000) {
  // vision calls commonly need 30s+: raise timeoutMs or set 0 to disable
}

Try / catch

try {
  await inspectImageTool.execute(id, params, signal);
} catch (e) {
  if (e instanceof ToolError && e.message.includes("timed out")) {
    // increase inspect_image.timeoutMs and retry, or check provider latency
  } else throw e;
}

Prevention

Set inspect_image.timeoutMs generously (60s+) for vision workloads; 0 disables.
Keep images small to reduce upload and inference latency.
Avoid slow/rate-limited providers for the vision role.

When it happens

Trigger: Vision request exceeded inspect_image.timeoutMs (settings) — slow provider, very large image, reasoning-heavy vision model, or a very low timeoutMs value; also network stalls pushing latency past the limit.

Common situations: timeoutMs set to a small value like 5000 while the vision model takes 20-60s; slow rate-limited provider; huge downscaled image increasing upload+inference time; thinking-enabled vision models adding latency.

Understand the failure class

Timeouts: ETIMEDOUT, deadlines, and hung requests — what actually expires when a request times out.

Related errors

AI-assisted analysis of can1357/oh-my-pi@9690622007 (2026-08-31). Data as JSON: /api/errors/402d5faa3f697776. Report an issue: GitHub.