build-report

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

• Third-party content exposure detected (high risk: 0.80). The skill explicitly ingests untrusted, user-provided build output (see SKILL.md Execution Phase
• Parse Step: "Input: Raw build output (string)" and sub-skills/parse-build-output.md), parses and interprets those messages, and then uses them to drive analysis and generate prioritized recommendations/commands, so arbitrary content in those logs could carry instructions that materially influence the agent's decisions and tool use.