st-delist-risk-scanner
Pass
Audited by Gen Agent Trust Hub on Sep 24, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructs the agent to execute Python scripts located in a sibling directory (
../findata-toolkit-cn/scripts/views_runner.py) to fetch financial data and manage the environment. These commands are localized to the repository structure and used for the skill's stated purpose.- [EXTERNAL_DOWNLOADS]: The environment preparation steps inreferences/data-queries.mdinvolve usingpipto install dependencies from arequirements.txtfile. This is a standard operation for Python-based agent tools.- [INDIRECT_PROMPT_INJECTION]: The skill processes financial data from external sources such as Baidu and EastMoney, which could potentially contain adversarial content, though the risk is low for this specific use case. - Ingestion points: External market data, news, and trading notifications fetched via data queries documented in
references/data-queries.md. - Boundary markers: The skill lacks explicit boundary markers or instructions to the LLM to ignore potentially injected instructions within the financial data.
- Capability inventory: The agent can execute shell commands through the Python interpreter to run data fetching scripts.
- Sanitization: No explicit sanitization or input validation logic is present in the instructions to handle untrusted data from the external APIs.
Audit Metadata