release

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The skill explicitly instructs fetching secrets via op and then embedding them verbatim into commands/flags (e.g., --otp=<otp>, piping "$PASSWORD", and NPM_CONFIG_TOKEN=<token>), which requires the agent to handle and output secret values directly.

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

• Third-party content exposure detected (high risk: 0.75). The required workflow reads OUTSIDER-authored free text via GitHub CLI commands that fetch PR/issue content from outside the operating user’s authorship (e.g., gh pr view <number> ... at SKILL.md:19 and git log/linking content at SKILL.md:17-23), then formats it into markdown/changelog that is fed back into the release context.