The Agent Skills Directory

[DATA_EXFILTRATION]: The skill drives a CLI tool that features a --deliver webhook:<url> sink, which allows the agent to POST command output—including sensitive identity data, tokens, and audit logs—to arbitrary external URLs.- [EXTERNAL_DOWNLOADS]: The skill requires the installation of the authentik-pp-cli binary and an MCP server, which are downloaded from the vendor's GitHub repository (github.com/mvanhorn/printing-press-library) using go install or npx.- [COMMAND_EXECUTION]: The skill executes the authentik-pp-cli tool to perform sensitive administrative operations, such as account recovery, user impersonation, token retrieval, and policy modification.- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection as it processes identity resource data from the Authentik API that could contain attacker-controlled instructions.
Ingestion points: Data retrieved from the Authentik API via core, flows, and events commands.
Boundary markers: The skill does not implement delimiters to separate API data from the agent's instructions.
Capability inventory: Includes high-privilege identity management and the ability to exfiltrate data via webhooks.
Sanitization: No specific filtering or sanitization of external API data is described.

pp-authentik