tao-generate-image-grounding

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 0.90). The prompt instructs the agent to collect API keys (e.g., vlm.gemini.api_key or api_key for OpenAI-compatible endpoints) and to place them into configuration/CLI overrides, which can require the LLM to accept and emit secret values verbatim.

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

• Third-party content exposure detected (high risk: 0.85). The skill ingests outsider-authored free text from the user-provided data.input_jsonl field caption (JSONL lines are arbitrary objects supplied at runtime), and that caption text is then sent into the VLM during Step 0 for expression extraction, so it can contain indirect prompt-injection content.