llm-wiki

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 0.90). The skill prompt instructs the agent to log into Obsidian using an explicit password passed on the command line (ob login --email <email> --password '<password>'), which exposes user credentials in command-line arguments and prompts the LLM to handle secret values directly.

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

• Third-party content exposure detected (low risk: 0.10). The agent ingests articles, papers, transcripts, and pasted text provided by the user or extracted from URLs, which constitute low-risk outsider-authored free text when fetched or supplied.