implement-task

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 0.90). The prompt requires the agent to paste the exact commands and command output into project files (including "the command" verbatim), which forces inclusion of any secrets or tokens that appear in those commands or outputs and thus creates an exfiltration risk.

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

• Third-party content exposure detected (high risk: 0.70). SKILL.md instructs the agent to read the “task packet” and “linked spec” from explicit full paths provided in the dispatch prompt, which are outsider-authored documents, and those files’ contents become LLM-readable context at runtime.