{
  "skill_name": "marketing-psychology",
  "skill_path": "/workspace/skills/coreyhaines31-marketingskills/skills/marketing-psychology",
  "skill_directory": "/workspace/skills/coreyhaines31-marketingskills/skills/marketing-psychology",
  "is_safe": true,
  "max_severity": "MEDIUM",
  "findings_count": 1,
  "findings": [
    {
      "id": "llm_finding_marketing-psychology_0",
      "rule_id": "LLM_PROMPT_INJECTION",
      "category": "prompt_injection",
      "severity": "MEDIUM",
      "title": "Transitive Trust Abuse via External Product Marketing Context File",
      "description": "The skill instructions direct the agent to read and trust content from `.claude/product-marketing-context.md` without validation. This file is external to the skill package and could contain malicious instructions that override the skill's intended behavior. The instruction 'Check for product marketing context first: If `.claude/product-marketing-context.md` exists, read it before applying mental models. Use that context to tailor recommendations' creates a transitive trust relationship where the skill delegates authority to untrusted external content.",
      "file_path": "SKILL.md",
      "line_number": null,
      "snippet": "Check for product marketing context first:\nIf `.claude/product-marketing-context.md` exists, read it before applying mental models. Use that context to tailor recommendations to the specific product and audience.",
      "remediation": "Remove the instruction to automatically read external context files, or implement strict validation if external context is necessary. Consider: (1) Bundling context files within the skill package instead of reading from external locations, (2) Adding explicit warnings that external files should be user-verified, (3) Implementing content validation before processing external files, (4) Requiring user confirmation before reading external context files.",
      "analyzer": "llm",
      "metadata": {
        "model": "claude-sonnet-4-5-20250929",
        "overall_assessment": "The marketing-psychology skill is primarily an educational resource providing mental models and psychological principles for marketing applications. The skill contains no executable scripts, no network operations, no data exfiltration attempts, and no command injection vulnerabilities. However, it exhibits one MEDIUM severity security concern: transitive trust abuse through automatic reading of external context files. The instruction to read `.claude/product-marketing-context.md` creates an indirect prompt injection vector where malicious content in that external file could override the skill's intended behavior. Additionally, the skill has minor documentation gaps with missing optional metadata fields (LOW severity, informational only). The core content\u2014mental models and marketing principles\u2014appears legitimate and educational with no signs of social engineering or deceptive intent.",
        "primary_threats": [
          "Indirect Prompt Injection (Transitive Trust)"
        ],
        "aitech": "AITech-1.2",
        "aitech_name": "Indirect Prompt Injection",
        "aisubtech": null,
        "aisubtech_name": null,
        "scanner_category": "PROMPT INJECTION",
        "meta_validated": true,
        "meta_confidence": "HIGH",
        "meta_confidence_reason": "The LLM analyzer correctly identified a transitive trust pattern. The instruction explicitly directs the agent to read an external file and 'use that context to tailor recommendations' without any validation or sandboxing. This creates an indirect prompt injection vector where malicious content in `.claude/product-marketing-context.md` could override intended behavior. The file path is outside the skill package (`.claude/` directory), making it untrusted content.",
        "meta_exploitability": "Medium - Requires attacker to place malicious content in `.claude/product-marketing-context.md` in the user's workspace. Not remotely exploitable, but feasible in shared environments or if user downloads malicious workspace files.",
        "meta_impact": "Medium - Could allow behavioral override of the skill's intended educational purpose. Malicious instructions in the context file could redirect the agent to perform unintended actions, though limited by the agent's overall safety constraints."
      }
    }
  ],
  "scan_duration_seconds": 17.55440664291382,
  "duration_ms": 17554,
  "analyzers_used": [
    "static_analyzer",
    "behavioral_analyzer",
    "llm_analyzer",
    "trigger_analyzer",
    "meta_analyzer"
  ],
  "timestamp": "2026-02-06T04:20:53.708245"
}