interrogate

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 0.90). This skill explicitly instructs the agent to package diffs/file contents and send them verbatim to spawned reviewers (via the Task tool/template), so any secrets present in those files would be included in the LLM's inputs/outputs.

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

• Third-party content exposure detected (medium risk: 0.30). The workflow ingests outsider-authored free text when it uses the user’s message (and PR description/commit messages if they exist) to derive the intent and then packages that intent with the diff for adversarial review.