caveman-compress

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The skill explicitly requires copying code blocks, inline code, URLs, commands, file paths, and environment variables "EXACTLY" which forces the model to reproduce any embedded API keys, tokens, or passwords verbatim if they appear in the input files.

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

• Third-party content exposure detected (medium risk: 0.30). In compress_file() the required workflow reads the user-specified local file (original_text = filepath.read_text(...)), then builds an LLM prompt containing that file’s free-form content (call_claude(build_compress_prompt(body))), so outsider-authored text is ingested whenever it is present in a file the user points the skill at.