The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes shell commands to perform environment detection, dependency analysis, and project validation. Examples include invoking npx tsc, node -v, and npm test to verify project state. The diagnostic script scripts/ts_diagnostic.py also uses subprocess.run with shell=True to execute hardcoded diagnostic commands such as grep and tsc. These executions are limited to one-shot diagnostics and are typical for developer tooling.
[INDIRECT_PROMPT_INJECTION]: The skill acts as an analysis engine for local codebases, creating a surface for indirect prompt injection via untrusted project files. Ingestion points: The skill reads and parses package.json and tsconfig.json, and executes grep across source files in the src/ directory. Boundary markers: Absent. The instructions do not define explicit delimiters or warnings to ignore instructions found within analyzed code or configuration files. Capability inventory: The skill can execute shell commands (subprocess.run, npm, npx) based on its internal logic and diagnostic results. Sanitization: The skill uses standard JSON parsing for configuration files, which ensures basic structure, but does not provide specific sanitization for downstream prompt interpolation.

typescript-expert