The Agent Skills Directory

[COMMAND_EXECUTION]: The skill provides instructions for the agent to use the Bash tool for project analysis tasks, such as mapping directory structures with tree, searching for files with find, and identifying patterns with grep. It also suggests using development CLIs like madge for dependency mapping and knip for finding unused code.
[DYNAMIC_EXECUTION]: The skill provides multiple TypeScript templates (e.g., analyze-structure.ts, analyze-deps.ts, move-module.ts) intended to be written and executed by the agent to perform complex file system operations and automated refactoring.
[INDIRECT_PROMPT_INJECTION]: The skill's primary function involves reading and analyzing untrusted source code from the ./src directory.
Ingestion points: Project files are ingested via Read, Grep, and Glob tools, as well as Node.js fs.readFileSync calls within provided scripts.
Boundary markers: The skill does not define specific delimiters or security instructions for the agent to ignore or isolate instructions potentially found within the analyzed data.
Capability inventory: The agent has significant capabilities, including local shell command execution and file system write operations.
Sanitization: There is no evidence of content sanitization or validation, meaning malicious instructions embedded in project files (such as in code comments or strings) could potentially influence the agent's behavior during analysis.

architecture-review