modlishka

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). This prompt instructs running a phishing reverse-proxy that captures and displays credentials and session cookies (and bypasses MFA), which requires the agent to collect and output secret values verbatim, creating high exfiltration risk.

CRITICAL E005: Suspicious download URL detected in skill instructions.

• Suspicious download URL detected (high risk: 1.00). The GitHub URL points to a repository (Modlishka) that is explicitly a reverse‑proxy phishing framework intended to capture credentials and bypass MFA (highly malicious if used), while accounts.google.com is a legitimate target site and 127.0.0.1:8888 is a local operator panel—together this indicates a high-risk, malicious distribution/use case despite GitHub itself being a reputable host.

CRITICAL E006: Malicious code pattern detected in skill scripts.

• Malicious code pattern detected (high risk: 1.00). The content explicitly documents Modlishka, a reverse-proxy phishing framework intended to perform transparent MITM attacks that capture credentials, session cookies, and bypass 2FA/MFA and includes operator/configuration for directing harvested data to an attacker-controlled domain—demonstrating clear, deliberate malicious intent (credential theft and data exfiltration).

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

• Third-party content exposure detected (high risk: 1.00). The skill's SKILL.md explicitly directs the agent to proxy and ingest live content from a real third-party site via the -target flag (e.g., https://accounts.google.com) and to apply jsRules and capture credentials in the observer panel, meaning untrusted public site content is fetched and interpreted as part of its workflow.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill's Quick Start instructs fetching and building remote code via "git clone https://github.com/drk1wi/Modlishka" and then running the resulting binary, so the external GitHub repository is fetched at runtime and executes remote code which the skill depends on.