The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill uses npx to execute the yandex-direct-api-mcp package. This package is retrieved from the NPM registry, a well-known service, and is a resource provided by the skill author.\n- [COMMAND_EXECUTION]: The skill utilizes the yandex-direct-mcp command-line interface to perform account management tasks. This involves the execution of shell commands to interact with the Yandex Direct API.\n- [INDIRECT_PROMPT_INJECTION]: The skill processes various text fields from the Yandex Direct API, such as campaign names and ad content.\n
Ingestion points: The agent reads data from methods like campaigns.get and reports.get listed in the methods and reports references.\n
Boundary markers: No explicit instructions are provided to the agent to treat API responses as inert data, which could allow instructions embedded in those responses to be executed.\n
Capability inventory: The skill enables the agent to perform sensitive actions like suspending campaigns, modifying bids, and deleting account resources.\n
Sanitization: There is no evidence of data validation or sanitization for strings returned by the API before they are presented to the agent.

yandex-direct