earnings-recap

Pass

Audited by Gen Agent Trust Hub on Sep 25, 2026

Risk Level: SAFEDYNAMIC_CONTEXT_INJECTIONCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADSDYNAMIC_EXECUTION
Full Analysis
  • [DYNAMIC_CONTEXT_INJECTION]: The skill utilizes the !command syntax within SKILL.md to execute a Python one-liner (python3 -c "exec(...)") when the skill is accessed. This code verifies the presence and version of the required yfinance library.\n- [COMMAND_EXECUTION]: The skill provides logic to automatically install the yfinance package using pip via subprocess.check_call if it is missing from the system.\n- [DYNAMIC_EXECUTION]: The environment check implemented in the dynamic context injection block uses the Python exec() function to run code from a string literal.\n- [EXTERNAL_DOWNLOADS]: The skill is designed to download the yfinance library from official repositories and fetch stock market data from Yahoo Finance's public services. These downloads are associated with the primary functionality of the skill.\n- [INDIRECT_PROMPT_INJECTION]: The skill ingests and displays stock news headlines through the ticker.news attribute. This introduces a surface where an external party could attempt to influence the agent by embedding instructions within news headlines.\n
  • Ingestion points: The ticker.news data source referenced in SKILL.md (Step 2) and Step 5.\n
  • Boundary markers: No explicit delimiters or instructions are used to separate the external news content from the agent's core logic.\n
  • Capability inventory: The skill allows for the execution of Python scripts and shell commands (via the installation step).\n
  • Sanitization: The skill does not perform explicit sanitization or filtering of the news headlines before they are presented to the agent.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 25, 2026, 10:32 PM