The Agent Skills Directory

[COMMAND_EXECUTION] (LOW): The script scripts/ts_diagnostic.py utilizes subprocess.run(shell=True) to execute system commands like grep, node, and npx. While the command strings are currently hardcoded, this is a security best-practice violation. Furthermore, running npx or tsc in an untrusted workspace can lead to local code execution if the workspace contains malicious configuration or hidden dependencies.\n- [PROMPT_INJECTION] (LOW): The skill is susceptible to indirect prompt injection (Category 8). \n
Ingestion points: scripts/ts_diagnostic.py reads and outputs contents from package.json, tsconfig.json, and performs recursive searches on the src/ directory. \n
Boundary markers: The skill lacks delimiters or explicit warnings to the agent when outputting file content or search results. \n
Capability inventory: The script can execute arbitrary shell commands and read local files. \n
Sanitization: There is no sanitization or filtering of the data read from the filesystem, meaning malicious instructions embedded in project files could influence the agent's behavior.

typescript-expert