The Agent Skills Directory

[REMOTE_CODE_EXECUTION]: The skill instructions include a command to download and execute an installation script for the uv package manager from astral.sh using a shell pipe (curl -LsSf https://astral.sh/uv/install.sh | sh). This source is recognized as a well-known technology provider.- [COMMAND_EXECUTION]: The skill relies on executing the aipa-cli tool, a vendor-specific utility, to retrieve financial ratios and company information.- [DYNAMIC_EXECUTION]: The workflow incorporates shell pipes that pass CLI output into Python one-liners (python3 -c) to compute financial metrics like sector averages and DuPont ROE decomposition. These scripts are provided as static templates in the skill instructions.- [INDIRECT_PROMPT_INJECTION]: The skill ingests and processes financial data from external sources, which constitutes a vulnerability surface for indirect prompt injection.
Ingestion points: Data retrieved from aipa-cli tool and web search results for international stocks and cryptocurrencies.
Boundary markers: The skill uses a structured 11-question analysis framework, but lacks explicit delimiters to isolate untrusted data.
Capability inventory: The agent can execute shell commands, run Python scripts, and perform web searches.
Sanitization: No explicit sanitization or validation of the retrieved financial data is documented before it is processed by the agent.

aipa-fundamentals