The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to automatically execute a shell command (ls) to inspect various skill directories (e.g., ~/.claude/skills/, ~/.agent/skills/) upon activation. This is used to check for the presence of recommended companion skills and offer installation links.
[COMMAND_EXECUTION]: The primary purpose of the skill is to facilitate the execution of arbitrary shell commands and git operations across multiple local directories using the meta exec and meta git sub-commands.
[EXTERNAL_DOWNLOADS]: The documentation provides instructions for installing the meta utility and its associated plugins globally via the NPM registry.
[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it reads and processes external data (the .meta configuration file and child repository content) that can influence which commands are executed.
Ingestion points: .meta configuration file and child repository paths (SKILL.md).
Boundary markers: Absent; no specific delimiters or warnings are provided to the agent to treat repository-sourced data as untrusted.
Capability inventory: Arbitrary shell command execution via meta exec, meta npm, and meta yarn (SKILL.md, references/commands.md).
Sanitization: Absent; the skill does not instruct the agent to validate or sanitize the contents of the configuration or the resulting commands.

meta-repo