The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes the Bash tool to execute various development and automation commands, including git operations for branch management (fetch, rebase, and dynamic HEAD resolution) and package manager scripts (bun run, bun test, npm audit) for code validation.\n- [INDIRECT_PROMPT_INJECTION]: The skill possesses an ingestion surface for indirect prompt injection by reading local repository files and capturing CLI tool outputs to determine the success or failure of the validation stage.\n
Ingestion points: The skill reads CODING_STANDARDS.md and ingests the raw output from git, bun, npm, and automated code review tools.\n
Boundary markers: There are no explicit instructions for the agent to use XML-like tags or specific delimiters when processing these external data sources.\n
Capability inventory: The skill is scoped to Bash, Read, Grep, and Glob tools, providing it with standard file system and command execution permissions within the development workspace.\n
Sanitization: The instructions do not require the agent to sanitize or filter content read from the repository or tool outputs before processing it.

validate