journalist-fit-check

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

• Third-party content exposure detected (high risk: 0.85). The skill’s runtime workflow requires fetching and reading outsider-authored free text from public web sources (e.g., host-agent-search results like author pages, Google News pages, and opened articles/bylines) to find anchor articles and verify bylines, which then gets ingested into the agent’s LLM context for analysis.

MEDIUM W021: Hidden or invisible Unicode characters detected (potential obfuscation or prompt injection).

• Hidden Unicode characters detected (1 type(s) found)