huawei-cloud-find-skills
Pass
Audited by Gen Agent Trust Hub on Jul 21, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill fetches a skill index and keyword mapping data from the GitCode API using the provided Python search script. It also retrieves skill documentation from GitHub for user review.
- [COMMAND_EXECUTION]: The skill requires the execution of a Python script (search-skills.py) to process search queries and uses npx to install selected skills onto the system.
- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection due to the ingestion of external data from remote repositories. 1. Ingestion points: The skill fetches dynamic JSON and Markdown content from gitcode.com and github.com via the search script and documentation fetch step. 2. Boundary markers: No specific delimiters or instructions are provided to the agent to disregard instructions within the fetched skill documentation. 3. Capability inventory: The agent can execute shell commands via Python and install/run remote packages via npx. 4. Sanitization: Content is parsed as JSON/Markdown but is not filtered for potential natural language instruction-based attacks.
Audit Metadata