The Agent Skills Directory

[EXTERNAL_DOWNLOADS] (MEDIUM): The skill instructs the agent to install the @ast-grep/cli package globally via npm. This involves downloading and installing third-party software from a non-whitelisted source at runtime.
[COMMAND_EXECUTION] (LOW): The skill relies on the agent's ability to execute shell commands, including npm install, ast-grep init, and ast-grep run. While necessary for the tool's function, these represent a capability surface that can be abused.
[PROMPT_INJECTION] (HIGH): High vulnerability to Indirect Prompt Injection (Category 8).
Ingestion points: The skill is designed to ingest and parse external codebases from the local filesystem (SKILL.md).
Boundary markers: There are no explicit delimiters or instructions to the agent to ignore natural language commands found within the code comments or strings of the analyzed files.
Capability inventory: The agent is granted shell command execution capabilities to install and run analysis tools.
Sanitization: No sanitization or filtering is performed on the content of the codebase before it is presented to or processed by the agent, allowing malicious instructions in the code to influence the agent's reasoning.

ast-grep-code-analysis