The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process local project configuration files, which can act as a vector for malicious instructions if the project files are sourced from an untrusted environment.
Ingestion points: Processes critical configuration files including package.json, tsconfig.json, jest.config.*, and .eslintrc.* (SKILL.md).
Boundary markers: No specific delimiters or "ignore embedded instructions" markers are established to prevent the agent from following instructions potentially hidden in these data files.
Capability inventory: The skill instructs the agent to execute powerful shell commands such as tsc, eslint, vitest, npm, and npx, which operate directly on the content of the ingested files (SKILL.md, REFERENCE.md).
Sanitization: There is no evidence of content validation or sanitization for the project files before they are processed by the agent or passed to CLI tools.
[COMMAND_EXECUTION]: The workflow relies on the execution of various development CLI tools including tsc --noEmit, eslint --fix, and vitest for project verification. These are standard operations for development-focused skills.
[EXTERNAL_DOWNLOADS]: The skill references standard dependency management operations such as npm ci and husky install (REFERENCE.md). These operations fetch code from the official npm registry, which is a well-known and trusted service for development workflows.

typescript-tooling