plan-analysis

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 0.90). The prompt requires the agent to "show actual code" and record currentContent for files (via grep/reads and evidence chains), which can force the LLM to reproduce file contents verbatim — including any embedded API keys, tokens, or passwords — creating an exfiltration risk.

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

• Third-party content exposure detected (high risk: 0.80). The skill explicitly reads and parses a user-provided implementation plan file ("Use when the user provides an implementation plan file" and phases describing "Read and parse the implementation plan"), meaning it ingests untrusted user-generated content that the agent will interpret.