privesc-linpeas

CRITICAL E006: Malicious code pattern detected in skill scripts.

• Malicious code pattern detected (high risk: 1.00). This repository is a purpose-built post-exploitation privilege-escalation toolkit with explicit, actionable instructions for credential harvesting, remote code execution, persistence (cron/systemd/SUID/ passwd modification), container breakout (docker.sock, nsenter), evasion techniques (curl|sh, tmpfs, clearing history), and a runner that fetches and executes code from arbitrary URLs — all patterns that enable deliberate unauthorized compromise and backdoor behavior.

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (high risk: 0.85). Outsider free text can be fetched at runtime from a public URL (--url defaulting to GitHub releases) via fetch_linpeas() and then executed/parsed by run_linpeas()/parse_output(), so LinPEAS output (attacker-controlled or tampered script content) becomes LLM-readable context through the runner’s generated findings/report text.

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

• Potentially malicious external URL detected (high risk: 1.00). High-confidence: the provided runner fetches and executes remote code at runtime from https://github.com/peass-ng/PEASS-ng/releases/latest/download/linpeas.sh (and the Quick Start shows curl ... | sh), so the skill depends on downloading and executing a remote script that directly runs code.

MEDIUM W013: Attempt to modify system services in skill instructions.

• Attempt to modify system services in skill instructions detected (high risk: 1.00). The prompt explicitly instructs running privilege-escalation and post-exploitation actions that modify system state (e.g., checking and exploiting sudo/SUID, adding a root user if /etc/passwd is writable, modifying system files, kernel exploits, shell escape techniques and evasion), thereby guiding compromise of the host.