The Agent Skills Directory

[COMMAND_EXECUTION]: The skill explicitly instructs the agent to execute shell commands using the npx ctx7 utility for all management tasks, forbidding the agent from merely printing the commands for user review.
[REMOTE_CODE_EXECUTION]: The install and info commands facilitate the retrieval and installation of code from external repositories. Specifically, npx ctx7 skills install allows for downloading third-party code into directories used by other AI agents (e.g., .claude/skills/, .cursor/skills/). While the skill implements a network permission gate, it does not provide a mechanism to verify the security or integrity of the remote content being installed.
[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from CLI search results to drive subsequent installation actions, which could be exploited by malicious repository entries.
Ingestion points: Data retrieved from npx ctx7 skills search and npx ctx7 skills info commands.
Boundary markers: The skill lacks explicit delimiters or instructions to ignore potential commands embedded within the search result output.
Capability inventory: The skill possesses the capability to execute shell commands (npx) and write to application-specific configuration directories.
Sanitization: There is no rigorous sanitization or validation logic defined for the content or URLs of the remote repositories beyond basic input correction.

context7-skills