external-cannbot-ops-ascendc-docs-search

Pass

Audited by Gen Agent Trust Hub on Sep 7, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill retrieves content from external web pages (Huawei Ascend community) to provide documentation to the agent.
  • Ingestion points: scripts/ascend_content_fetcher.py and scripts/ascend_search_skill.py fetch data from www.hiascend.com.
  • Boundary markers: The fetched content is returned to the agent within structured JSON objects.
  • Capability inventory: The skill utilizes network requests (GET) and has file manipulation capabilities through the provided scripts/clean_markdown.py.
  • Sanitization: Content is sanitized using _clean_text in scripts/ascend_content_fetcher.py to remove scripts and styles, and Markdown cleanup is performed by scripts/clean_markdown.py.
  • [EXTERNAL_DOWNLOADS]: The skill communicates with www.hiascend.com to perform documentation searches and fetch page content. This is directed at a well-known service for developer documentation.
  • [COMMAND_EXECUTION]: The provided scripts/clean_markdown.py utility handles file reading and writing for the purpose of cleaning up Markdown files. While intended for documentation maintenance, it grants the agent the ability to modify Markdown files within its accessible directory.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 7, 2026, 03:53 PM
Security Audit — agent-trust-hub — external-cannbot-ops-ascendc-docs-search