from claude-matrix108
Manage file and package warnings (add, remove, list, check) for a Matrix workflow; parse user arguments and call the matrix_warn tool to enforce repository-leve
Matrix Warnings lets an agent manage and query warnings for problematic files or packages within a repo or workspace. It parses simple user commands (add, remove, list, check) and invokes the matrix_warn tool to implement actions such as repo-scoped warnings, severity levels, and targeted checks. The skill provides examples and expected parameter structures for each action.
Use this skill when a user asks to add or remove warnings, list current warnings, check a file or package for warnings, or otherwise manage "grudges" that should block or flag problematic code or dependencies. Ideal for code quality workflows, deprecation enforcement, and repository-specific policy checks.
type, target, reason, and severity parameters when calling the matrix_warn tool.Designed for agents that can call platform tools/plugins (e.g., Claude Code agents integrated with matrix_warn). Works best where a matrix_warn tool/plugin is available to perform the backend operations.
A thin skill that wraps a single Matrix plugin tool (matrix_warn) to manage file/package warnings ('grudges'). No scripts included — purely instructional SKILL.md that maps user phrases to tool actions. Functional but minimal: no error handling guidance, no edge cases, no validation. Only useful for users already running the specific Matrix MCP plugin.
Niche skill for a specific Matrix workflow plugin. Clean from a security standpoint — no scripts, no network calls, no credential handling. Architecture is flat with no separation of concerns. Usefulness is limited to Matrix plugin users.
