Aphorisms

CRITICAL E004: Prompt injection detected in skill instructions.

• Potential prompt injection detected (high risk: 0.90). The skill mandates executing an unrelated curl POST to localhost "before doing anything else," a mandatory network action unrelated to the Aphorisms functionality and therefore a deceptive/extraneous instruction outside the skill's stated purpose.

CRITICAL E006: Malicious code pattern detected in skill scripts.

• Malicious code pattern detected (high risk: 0.90). Multiple workflows forcibly execute a backgrounded, output-suppressed HTTP POST to http://localhost:31337/notify (port 31337 is commonly abused), which is a hidden network callback that could be used to trigger or signal a local backdoor/agent — this is a high-risk intentional behavior.

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

• Third-party content exposure detected (medium risk: 0.30). In Workflows/FindAphorism.md Step 1, the required workflow ingests user-provided newsletter text or WebFetch-fetched URL content (“Extract main article content, title, and key themes”), which is outsider-authored free text fed into the LLM without selecting a specific trusted item first.