weibo-comment-search
Pass
Audited by Gen Agent Trust Hub on Aug 15, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes local Python scripts weibo_comment_search.py and backfill_html.py to handle data retrieval and template population. It also utilizes the open command to display generated reports to the user.
- [EXTERNAL_DOWNLOADS]: The skill retrieves data from https://redfox.hk which is the official API domain for the skill vendor. This connection is used to fetch the Weibo comment data required for the skill's primary function.
- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted Weibo comments that are subsequently processed by the AI for summarization, creating a potential surface for indirect prompt injection. (1) Ingestion points: Comments are fetched via API in weibo_comment_search.py and presented in the agent's context. (2) Boundary markers: The instructions do not define explicit delimiters or instructions for the agent to ignore embedded commands within the comments. (3) Capability inventory: The skill possesses the ability to execute local Python scripts, write files to the local Downloads directory, and open files in a web browser. (4) Sanitization: The skill includes functions for escaping content for Markdown tables and encoding HTML entities for report generation.
Audit Metadata