The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill instructs the agent to install a global Node.js package @aisuite/chub if it is not already present in the environment. This dependency originates from an organization that is not recognized as a trusted source, presenting a supply chain risk where the package could contain malicious code.
[COMMAND_EXECUTION]: The skill requires the execution of the chub CLI and its associated subcommands to search for and retrieve documentation. Executing binaries from an unverified external package allows for arbitrary command execution within the agent's shell environment.
[INDIRECT_PROMPT_INJECTION]: The skill fetches documentation from an external source for processing, which is an attack surface for indirect prompt injection. Malicious instructions hidden within the documentation could be used to hijack the agent's logic.
Ingestion points: Documentation content retrieved via chub get <id> is incorporated into the agent's context.
Boundary markers: There are no defined delimiters or safety instructions in the skill to isolate external content from the agent's primary instructions.
Capability inventory: The agent has the ability to execute shell commands and generate executable code, which could be misused if the documentation contains malicious prompts.
Sanitization: The skill does not implement any validation or sanitization of the content fetched from the external documentation service.

get-api-docs