The Agent Skills Directory

[PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface as it processes external content to drive its automation logic.
Ingestion points: The agent reads potentially untrusted text from Linear issue descriptions, comments (via mcp__linear-server__get_issue), and GitHub pull request diffs (via gh pr diff).
Boundary markers: No explicit delimiters or instructions are provided to the agent to treat external content as untrusted data or to ignore embedded instructions.
Capability inventory: The skill leverages Bash, Write, and Edit tools, enabling it to execute commands and modify the local codebase based on recommendations parsed from external sources.
Sanitization: There is no evidence of validation or sanitization applied to the text retrieved from external task management or version control systems.
[EXTERNAL_DOWNLOADS]: The skill retrieves data and code fragments from trusted and well-known services.
Downloads issue data and developer comments from the official Linear service.
Fetches pull request diffs and state information from the authoritative nrwl/nx repository on GitHub.
[COMMAND_EXECUTION]: The skill uses shell commands to verify compliance and interact with repository tooling.
Executes npx nx test to run plugin test suites after modifications.
Utilizes the GitHub CLI (gh) to query PR metadata and content.
Uses git for branch management and local diff evaluation during the review and implementation phases.

multi-version-compliance