The Agent Skills Directory

[REMOTE_CODE_EXECUTION]: Downloads and executes an installation script for the 'uv' package manager from the official astral.sh domain. This targets a well-known technology provider and is an established method for tool installation.
[DYNAMIC_EXECUTION]: The skill instructs the agent to generate and execute Python one-liners using 'python3 -c' to verify market metrics such as volume averages and R:R ratios. This practice involves runtime code generation and execution to process data results.
[INDIRECT_PROMPT_INJECTION]: The skill ingests external market data from S3 archives which is processed through CLI tools and Python logic. This ingestion of untrusted content creates a surface for potential injection if the data source were compromised. Evidence: (1) Ingestion points: Market data fetched via aipa-cli subcommands. (2) Boundary markers: No explicit delimiters or ignore-instructions warnings are present for the piped data streams. (3) Capability inventory: Includes local Python execution and file system access for data caching. (4) Sanitization: The provided Python templates use string splitting and casting to numeric types (int and float) to parse input rows.
[EXTERNAL_DOWNLOADS]: References 'pip install' and 'uvx' for downloading the 'aipa-cli' package. These are legitimate installation methods for the skill's primary functionality.

aipa-data