The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill fetches React best practice standards and rules from the official somnio-software GitHub repository (raw.githubusercontent.com) when local copies are unavailable. These documents are used as reference data for the audit and do not contain executable code.\n- [COMMAND_EXECUTION]: Executes local shell commands using the Bash tool to retrieve the project's repository name via git remote and to perform file system enumeration using find and wc. The results are used for report naming and architectural inventory.\n- [INDIRECT_PROMPT_INJECTION]: As an auditing tool, the skill reads and processes user-provided source code which could theoretically contain instructions designed to deceive the analyzer.\n
Ingestion points: Project files including .tsx, .ts, and .test.tsx are read for analysis by the scanner and analyzer agents.\n
Boundary markers: None. Instructions do not specify delimiters to separate code from instructions.\n
Capability inventory: The skill possesses Write access to the reports directory, Bash tool for local command execution, and Agent tool for sub-agent management.\n
Sanitization: No content sanitization is performed on the source code before it is processed by the analysis models.

react-best-practices