Information exposure alert through an exception

[arpitjain099]

Describe the bug

Stack trace information flows to this location and may be exposed to an external user.

Affected line -

evals/evals/elsuite/multistep_web_tasks/docker/flask-playwright/app.py

Line 191 in 234bcde

jsonify({"status": "error", "message": f"error executing command {command}: {e}"}),

evals/evals/elsuite/multistep_web_tasks/docker/flask-playwright/app.py

Lines 164 to 170 in 234bcde

	response = jsonify(
	{
	"status": "success",
	"message": f"could not return results of executed commands {request.json['commands']}",
	"content": str(e),
	"url": page.url,
	}

evals/evals/elsuite/multistep_web_tasks/docker/flask-playwright/app.py

Lines 119 to 124 in 234bcde

	response = jsonify(
	{
	"status": "success",
	"message": f"could not return results of executed commands {request.json['command']}",
	"content": str(e),
	"url": page.url,

evals/evals/elsuite/multistep_web_tasks/docker/flask-playwright/app.py

Lines 61 to 65 in 234bcde

	except Exception as e:
	return jsonify(
	{"status": "error", "message": f"failed to start session (already started?): {e}"}
	)
	return jsonify({"status": "success", "message": "session started"})

To Reproduce

evals/evals/elsuite/multistep_web_tasks/docker/flask-playwright/app.py

Line 191 in 234bcde

jsonify({"status": "error", "message": f"error executing command {command}: {e}"}),

evals/evals/elsuite/multistep_web_tasks/docker/flask-playwright/app.py

Lines 164 to 170 in 234bcde

	response = jsonify(
	{
	"status": "success",
	"message": f"could not return results of executed commands {request.json['commands']}",
	"content": str(e),
	"url": page.url,
	}

evals/evals/elsuite/multistep_web_tasks/docker/flask-playwright/app.py

Lines 119 to 124 in 234bcde

	response = jsonify(
	{
	"status": "success",
	"message": f"could not return results of executed commands {request.json['command']}",
	"content": str(e),
	"url": page.url,

evals/evals/elsuite/multistep_web_tasks/docker/flask-playwright/app.py

Lines 61 to 65 in 234bcde

	except Exception as e:
	return jsonify(
	{"status": "error", "message": f"failed to start session (already started?): {e}"}
	)
	return jsonify({"status": "success", "message": "session started"})

Code snippets

https://github.com/openai/evals/blob/234bcde34b5951233681455faeb92baaaef97573/evals/elsuite/multistep_web_tasks/docker/flask-playwright/app.py#L191-L191

https://github.com/openai/evals/blob/234bcde34b5951233681455faeb92baaaef97573/evals/elsuite/multistep_web_tasks/docker/flask-playwright/app.py#L164-L170

https://github.com/openai/evals/blob/234bcde34b5951233681455faeb92baaaef97573/evals/elsuite/multistep_web_tasks/docker/flask-playwright/app.py#L119-L124

https://github.com/openai/evals/blob/234bcde34b5951233681455faeb92baaaef97573/evals/elsuite/multistep_web_tasks/docker/flask-playwright/app.py#L61-L65

OS

macOS

Python version

3.11.4

Library version

1.40.1