The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to fetch and process real-time data from external sources using tools such as get_ticker_prices, get_simple_earn_config, get_nft_mining_rewards, and get_mining_income. This creates a vulnerability surface where the agent could potentially be influenced by instructions embedded within the data returned by these tools.
Ingestion points: Tool outputs from get_ticker_prices, get_simple_earn_config, get_nft_mining_rewards, and get_mining_income are ingested into the agent's context as live data.
Boundary markers: The instructions do not provide explicit delimiters or "ignore embedded instructions" warnings for the agent when processing these tool outputs.
Capability inventory: The skill is a knowledge-based extension and does not contain its own scripts, file-writing operations, or network-post capabilities; however, it leverages platform tools to aggregate information.
Sanitization: There are no requirements or instructions within the skill for the agent to sanitize, escape, or validate the content fetched from the external tools.

gomining-investment-guide