find-examples

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 0.90). The skill instructs the agent to search repositories and present verbatim "relevant code snippets" and config/usage examples, which could force the LLM to output hard-coded credentials or API keys found in those files, creating an exfiltration risk.

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

• Third-party content exposure detected (high risk: 0.75). The required workflow is to run org-wide code_search/search and then present “vetted examples,” which implies the agent ingests outsider-authored repository/documentation text (e.g., issues/PRs/wiki or other teams’ code/docs) into the LLM context for evaluation and summarization.