tiangong-kb-edu-search

Pass

Audited by Gen Agent Trust Hub on Aug 24, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The wrapper script uses npx to download and execute the @tiangong-ai/cli package from the NPM registry. This is a standard practice for managing the life cycle of the vendor-provided command-line interface tool.
  • [COMMAND_EXECUTION]: The edu_search.sh script executes shell commands to interact with the Tiangong AI CLI. It provides a flexible override mechanism via environment variables TIANGONG_AI_CLI and TIANGONG_AI_CLI_BIN, enabling the entry point to be customized for different execution environments.
  • [DATA_EXPOSURE]: The skill facilitates access to the local filesystem to load environment variables. It accepts an env_file path and also checks for a .env file in the directory of any provided request_file. This mechanism is used to manage authentication credentials and service configurations required for the search operation.
  • [INDIRECT_PROMPT_INJECTION]: * Ingestion points: Content retrieved from the knowledge base via the education search command is returned directly to the agent's context. * Boundary markers: No explicit delimiters or instructions to ignore embedded commands are present in the resulting output. * Capability inventory: The skill has the capability to execute shell commands and read files from the local filesystem. * Sanitization: There is no evidence of filtering or escaping of the retrieved content before it is processed by the agent.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 24, 2026, 05:00 PM
Security Audit — agent-trust-hub — tiangong-kb-edu-search