The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill is configured to fetch machine-readable markdown from https://clig.dev/llms.txt. This is a well-known documentation site for CLI development standards. The skill instructions also suggest using the tavily-extract or WebFetch tools to retrieve this data at runtime to ensure recommendations remain current.
[COMMAND_EXECUTION]: The skill contains examples and guidance involving command-line flags and arguments (e.g., handling secrets, exit codes, and standard streams). These are provided as static documentation for the agent to use when advising users and do not involve the execution of shell commands by the skill itself.
[DATA_EXFILTRATION]: The skill documentation correctly identifies security best practices regarding secrets, specifically advising against placing credentials in environment variables or command-line flags to avoid leakage. No patterns indicating unauthorized data access or transmission were found.
[PROMPT_INJECTION]: The skill provides structured instructions for the agent to act as a reviewer or designer based on external guidelines. It does not contain instructions that attempt to bypass safety filters or override the agent's core identity.

cli-development