The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted user input that is used to construct a shell command invocation. \n- Ingestion points: SKILL.md instructs the agent to collect values for investment amount, period, and CDI percentage from user messages. \n- Boundary markers: Absent. There are no delimiters or instructions to treat the data as literal strings or to prevent shell metacharacter injection. \n- Capability inventory: The skill uses scripts/cdi.py via a shell command execution capability. \n- Sanitization: Absent. The instructions do not specify validation or sanitization of user input before it is interpolated into the command line template. \n- [EXTERNAL_DOWNLOADS]: The script performs network requests to fetch current interest rates. \n- Evidence: scripts/cdi.py makes requests to the official Banco Central do Brasil API and the financial portal melhorcambio.com to fetch the CDI rate. \n- [COMMAND_EXECUTION]: The skill relies on executing a local script with variable arguments. \n- Evidence: SKILL.md contains a template for executing python scripts/cdi.py with arguments derived from user input.

calculadora-cdi