code-review

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The skill explicitly requires showing "Current code" snippets from git diffs and writing them into reports (including files), so any API keys, tokens, or passwords present in the diffs would be reproduced verbatim in the LLM output and persisted reports.

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

• Third-party content exposure detected (high risk: 0.90). The skill's workflow explicitly runs commands that fetch public, user-generated pull request diffs (e.g., "review PR #N" → gh pr diff N in SKILL.md), which the agent ingests and analyzes as part of its decision-making, exposing it to untrusted third‑party content from GitHub PRs.