The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructions require the agent to execute several shell commands using the autor3search-javascript CLI tool, including init, doctor, baseline, eval, and report.
[EXTERNAL_DOWNLOADS]: The skill performs a global installation of the @autor3search/javascript package via NPM. This package is managed by the skill's author and is central to the skill's performance optimization functionality.
[INDIRECT_PROMPT_INJECTION]: The skill creates a significant attack surface by instructing the agent to read and follow an external file (program.md) found in the user's repository as its primary instruction set.
Ingestion points: The program.md file located within the repository being optimized.
Boundary markers: None identified. The instructions explicitly command the agent to "Follow it exactly" without safety delimiters or warnings to ignore embedded instructions.
Capability inventory: The agent has the ability to execute shell commands (via the vendor CLI), perform Git operations (commit, hard reset), and modify project source code.
Sanitization: None identified. The agent is directed to treat the file content as a verbatim instruction set.

autor3search-javascript