The Agent Skills Directory

Command Execution: The skill invokes development tools including git, pnpm, and ember to perform its objective. These commands are used to build packages, lint source code, and execute test suites, which is consistent with the skill's primary purpose for repository maintenance.- Potential for Indirect Prompt Injection: The skill processes external data from the codebase to drive its logic, which introduces a potential surface for indirect instructions.
Ingestion points: The skill ingests data from the repository environment in SKILL.md using git merge-base, git log, and git diff-tree to identify changed files and commit metadata.
Boundary markers: There are no specific delimiters or instructions to treat repository-derived strings as untrusted when interpolating them into subsequent commands.
Capability inventory: The skill possesses the capability to execute shell commands across several directories (packages/components, packages/mcp, website, etc.) via the pnpm and ember tools.
Sanitization: The procedure does not explicitly mention sanitization or validation of the variables derived from the file system, such as {component-name} or {docs/changed-page-file-path}, before they are passed as arguments to shell execution strings in package-components-checks.md and website-checks.md.

run-checks