The Agent Skills Directory

[REMOTE_CODE_EXECUTION]: The documentation within SKILL.md provides a command that downloads a script from https://app.sikaru.ai/install.sh and pipes it to sh. This enables the execution of unverified remote code on the host system.
[COMMAND_EXECUTION]: The skill directs the agent to interact with the system via the sikaru CLI, including commands for setup, inference, and agent management.
[EXTERNAL_DOWNLOADS]: The skill facilitates the download and installation of packages from NPM and PyPI, and fetches documentation from the sikaru.ai domain.
[INDIRECT_PROMPT_INJECTION]:
Ingestion points: Reads external documentation from docs.sikaru.ai/llms.txt, OpenAPI schemas, and monitors run statuses from API events and the sikaru-docs MCP server.
Boundary markers: The instructions do not define delimiters or specific warnings to ignore instructions that might be embedded in the retrieved documentation or API data.
Capability inventory: The agent can execute shell commands, perform network requests, and manage local configurations.
Sanitization: There are no instructions provided for sanitizing or validating the content retrieved from external sources before it is used to determine subsequent actions.

sikaru