aeon-skill-security-scan

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The skill scans file contents and reports matches as {file, line, pattern, severity} (and prints matched lines in findings) but gives no instruction to redact secrets, so the LLM would need to read and emit secret values verbatim if present.

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

• Third-party content exposure detected (high risk: 0.85). The required workflow scans <skills-dir>/*/SKILL.md, companion scripts, and <skills-dir>/*/references/* and then feeds the matched findings (file/line/pattern context) into the agent’s LLM output, so it ingests outsider-authored free text from other installed skills’ documentation/scripts at runtime.