flux-opencode-plan

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

• Third-party content exposure detected (medium risk: 0.30). Source: SKILL.md input “Full request: $ARGUMENTS”; runtime path: Step 1-3 use that outsider-authored natural-language request as the passed into multiple research/gap-analysis subagents (flux:context-scout/repo-scout/practice-scout/docs-scout/github-scout/flux-gap-analyst), meaning the LLM ingests free text supplied by the workflow caller before selecting any specific item.