signoz

CRITICAL E004: Prompt injection detected in skill instructions.

• Potential prompt injection detected (high risk: 0.80). The "Companion check" section instructs the agent to run local shell commands to enumerate and offer installing companion skills—an action unrelated to SigNoz assistance that can expose local environment details and thus acts as an out-of-scope, potentially exfiltrative instruction.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 0.90). The setup instructions include runtime commands that fetch and execute remote content — e.g., "git clone -b main https://github.com/SigNoz/signoz.git" (cloning a repo then running docker compose from it) and "helm repo add signoz https://charts.signoz.io" (adding a chart repo and installing charts), which download and run remote code/configuration.