The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: Fetches and executes code from the author's GitHub repository timschoch/skilly using npx. This enables dynamic tool updates at runtime directly from the source repository.
[REMOTE_CODE_EXECUTION]: The skill instructs the agent to execute remote code via npx from both the NPM registry (skills package) and GitHub (timschoch/skilly). This allows for the execution of arbitrary scripts provided by those sources.
[COMMAND_EXECUTION]: The management process involves automated command-line operations, including file system writes to .claude/skills/ and .agents/skills/, and Git operations such as branching, committing, and pushing for automated PR creation.
[INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied URLs for skill installation, which presents an attack surface where malicious instructions could be embedded in the fetched content. Ingestion points: User-provided URLs via the npx skills add <url> command. Boundary markers: None present to delimit or warn the agent about potential instructions within the fetched skill content. Capability inventory: The skill has the ability to execute shell commands (npx), write to the filesystem, and perform Git operations. Sanitization: No evidence of sanitization or validation of the content fetched from external URLs before installation.

skilly-cli