linkedin-writer

Pass

Audited by Gen Agent Trust Hub on May 12, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill implements a robust, interactive workflow that requires user confirmation at every step of the content generation process.
  • [EXTERNAL_DOWNLOADS]: Accesses external content from YouTube and web articles via the Apify MCP and WebFetch tools. These actions are clearly defined as necessary for the skill's purpose of repurposing source material.
  • [DATA_EXFILTRATION]: No unauthorized data access or transmission was observed. The skill uses its internal reference files exclusively to maintain the requested persona and voice.
  • [PROMPT_INJECTION]: Instructions are focused on guiding the agent through a specific creative process and do not contain attempts to bypass safety protocols or extract system information.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external data (transcripts and web content). The potential for indirect prompt injection is mitigated by the mandatory human-in-the-loop workflow requiring user approval of outcomes and skeletons in Step 0, Step 2, Step 3, and Step 4 before final content is generated in an artifact.
Audit Metadata
Risk Level
SAFE
Analyzed
May 12, 2026, 03:09 AM