soia-dev-github-ops

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

• Third-party content exposure detected (high risk: 0.85). The required workflow’s pre-merge rule-review procedure pulls outsider-authored PR content (PR title/body and diff) via gh pr view/gh pr diff, which ingests free-form text authored by others into the agent context at runtime.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 0.90). The skill explicitly instructs at runtime to fetch raw repository rule files using "gh api repos///contents/?ref= --jq '.content' | base64 -d", which can inject repo-maintained rule text into the agent's context and directly control its prompt/behavior.