The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to 'run every repository-mandated full verification command' found in files such as AGENTS.md, manifests, or CI configuration. This is a high-risk capability because it grants the content of files within the workspace the authority to define and execute arbitrary shell commands.
[DYNAMIC_EXECUTION]: The skill executes a Node.js script (validate-llm-wiki.mjs) using a path that includes a variable for another skill's directory. Executing scripts from dynamic paths can be exploited if the path resolution or the script itself is compromised.
[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from the repository being reviewed, which could contain malicious instructions.
Ingestion points: External files including AGENTS.md, CI configurations, manifests, and any file content encountered during search and link-rewriting operations in the target repository.
Boundary markers: None identified. The skill does not provide instructions to the agent to delimit or treat external file content as untrusted.
Capability inventory: The skill can execute arbitrary shell commands (if found in configuration files), run specific Node.js scripts, perform git operations, search the filesystem, and move or modify files.
Sanitization: None identified. There is no instruction to validate or sanitize the commands extracted from repository files before they are executed.

review-and-verify