ui-update

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

• Third-party content exposure detected (low risk: 0.10). The skill’s runtime workflow clones and reads outsider-authored repository files from https://github.com/bitjaru/styleseed.git at a pinned commit (after explicit user approval), so it ingests free text from an external source authored by third parties.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill clones an upstream repository at runtime (git clone https://github.com/bitjaru/styleseed.git) and then inspects and copies agent-instruction files (e.g., CLAUDE.md, .claude/skills) from that repo to update local prompts, so the fetched content can directly control agent instructions.