The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses the aikidoctl events --json command to retrieve local security logs. This is a read-only diagnostic operation used to determine if a network request or package installation was blocked by the security agent.
[INDIRECT_PROMPT_INJECTION]: The skill ingests and parses JSON output from a local security tool that monitors external interactions (e.g., package registry requests). While the package names and registry responses contained in these logs could potentially include malicious strings designed to influence the agent's behavior, the skill's logic is focused on status reporting, which significantly limits the potential for automated misuse.
Ingestion points: The output of aikidoctl events --json is processed by the agent in the Check before you debug section of SKILL.md.
Boundary markers: The skill does not define specific delimiters or instructions to ignore embedded content when parsing the JSON output.
Capability inventory: The skill is limited to diagnostic reporting and does not possess capabilities for file writing, system modification, or outbound network requests based on the log data.
Sanitization: There is no explicit sanitization or filtering mentioned for the data fields retrieved from the event logs.

aikido-device-protection