morning-brief

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 0.90). The skill instructs the agent to extract and quote user message text from local transcripts and to produce "ready-to-paste" command lines, which could cause the LLM to include any secrets (API keys, tokens, passwords) found in those transcripts verbatim.

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

• Third-party content exposure detected (high risk: 0.70). The required workflow reads filesystem transcript JSONL files from ~/.claude/projects/*/*.jsonl and extracts their user message text into the agent’s context; those transcript contents are outsider-authored free-form messages (by other people than the operating user) if present.