The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to execute a shell command to resolve the path for library references using the CLAUDE_LIBRARY_DIR environment variable or a default path in the user's home directory. This is used solely for local resource discovery.- [DATA_EXFILTRATION]: The skill reads source code from user-defined project paths and accesses library reference files located in ~/.claude/library/. Access is limited to local files for the purpose of code analysis and no external network transmission of this data is programmed.- [PROMPT_INJECTION]: The skill processes untrusted source code which presents a surface for indirect prompt injection.
Ingestion points: Target source files (.ts, .tsx) identified via user-provided paths or glob patterns in Step 1.
Boundary markers: None present; the skill reads and processes the file content directly into context.
Capability inventory: The agent can perform shell command execution for path resolution and file system reads across the target project.
Sanitization: No sanitization or filtering of the ingested code content is performed before analysis. As the primary purpose of the skill is code analysis, this is a known characteristic of the workflow.

ts-audit