The Agent Skills Directory

[COMMAND_EXECUTION]: The diagnostic script scripts/ts_diagnostic.py utilizes subprocess.run with shell=True to execute system commands like tsc, node, and grep for project analysis. These operations are restricted to hardcoded diagnostic queries and standard project paths.
[COMMAND_EXECUTION]: In SKILL.md, the agent is instructed to use a node -e one-liner to parse the package.json file. This is a standard method for identifying project dependencies and tooling without requiring additional external libraries.
[EXTERNAL_DOWNLOADS]: The skill frequently invokes npx for tasks like type checking (npx tsc) and testing (npx vitest). These tools may download package updates from the npm registry if they are not already cached on the system.
[PROMPT_INJECTION]: The skill possesses an indirect prompt injection surface because it processes untrusted data from the user's project files during diagnostics. This is a common property of developer tools and is handled safely within the skill's scope.
Ingestion points: The skill reads and analyzes package.json, tsconfig.json, and TypeScript source files in the src/ directory (found in SKILL.md and scripts/ts_diagnostic.py).
Boundary markers: No specific boundary markers or delimiters are used to isolate project data from the instructions.
Capability inventory: The skill has the ability to execute shell commands and read/write files as part of its expert development functions.
Sanitization: No explicit sanitization or filtering of project file content was detected before the data is processed by the agent or diagnostic scripts.

typescript-expert