The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to create and manage git worktrees (git worktree add) to isolate the audit environment from the main branch. It also executes verification commands (test, build, lint) that are either provided by the user or inferred from the project's environment.
[INDIRECT_PROMPT_INJECTION]: The skill ingests and analyzes external TypeScript source files, which creates an attack surface for instructions hidden in comments or code to influence agent behavior.
Ingestion points: TypeScript files (.ts) within the targeted project directory.
Boundary markers: No explicit delimiters or boundary markers are defined in the instructions to separate source code from agent instructions.
Capability inventory: The skill has capabilities for file system operations (reading source and writing to .agents/), git repository manipulation, and execution of shell-based verification commands.
Sanitization: No explicit sanitization, escaping, or filtering of input file content is implemented before the agent processes the code.
[EXTERNAL_DOWNLOADS]: The skill documentation refers to downloading and installing the skill and its related dependencies (specialized audit skills) from the vendor's GitHub repository (github.com/gohypergiant/agent-skills). These references are recognized as legitimate vendor-owned resources.

accelint-ts-audit-all