The Agent Skills Directory

[REMOTE_CODE_EXECUTION]: The skill provides instructions to install the uv package manager by downloading a script from a well-known service (https://astral.sh/uv/install.sh) and piping it directly to the shell (sh).
[DYNAMIC_EXECUTION]: The agent is instructed to generate Python code snippets and execute them using python3 -c to perform calculations on data retrieved from the CLI, such as volume multipliers and risk-reward ratios.
[COMMAND_EXECUTION]: The skill requires the execution of multiple shell commands to function, including pip for package installation and uvx for running the aipa-cli tool.
[INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted external data from financial markets and news sources (e.g., CafeF, VNExpress) to perform its analysis.
Ingestion points: Data is brought into the agent's context via the aipa get-ohlcv-data command and external news research instructions in SKILL.md.
Boundary markers: The instructions do not specify the use of clear delimiters or boundary markers to separate untrusted external data from the agent's internal instructions.
Capability inventory: The skill utilizes shell command execution, Python script execution, and multi-agent task spawning.
Sanitization: There is no evidence of sanitization or filtering logic applied to the external data before it is processed by the agent pipeline.

aipa-research