Skip to main content
Cybersecurity Statistics

Cybersecurity statistics / Prompt Injection

Roughly 5% of published Skills in one marketplace were malicious or high risk.

PublisherStraiker
ReportThe Year Agents Entered the Workforce
Published14 July 2026
TopicsPrompt Injection, Agent Skills

Published by Straiker in The Year Agents Entered the Workforce, 14 July 2026. The figure is taken from the report as published; the full methodology is in the source.

View the original report

Related statistics

Detections of long, malicious prompt-injection payloads increased roughly fivefold between March and May 2026.
Check Point, 18/07/2026
48% of security teams report blind spots around prompt injection chains or tool-chaining abuse in AI-native applications.
Rein Security, 22/02/2026
More than 1,700 successful exploits occured across production coding, productivity, and first-party AI agents during adversarial testing.
Straiker, 18/07/2026
36% of successful attacks on coding agents reached remote code execution on the developer's machine.
Straiker, 18/07/2026
91% of successful attacks on productivity agents ended in silent data exfiltration.
Straiker, 18/07/2026
Nearly a quarter (24%) of 17,651+ tracked Model Context Protocol (MCP) servers carry at least one vulnerability.
Straiker, 18/07/2026

Get the newsletter

Weekly cybersecurity statistics by email.