VdustR

SKILL_DESCRIPTION - when to use this skill

Session retrospective — review recent work to discover improvement opportunities through interactive dialogue. Analyzes corrections, undocumented practices, efficiency patterns, tool usage, and workflow gaps. Use when the user wants to reflect on recent work and find ways to improve their AI collaboration workflow. Common moments: end of session, after a PR, after debugging, after a code review, after finishing a major task, or whenever something felt inefficient. Use when asked to "retro", "session retro", "session review", "review this session", "what can I improve", "retrospective", "what went wrong", "how can I be more efficient", or when the user wants to improve their CLAUDE.md, discover useful skills, optimize existing skills, or design new workflows based on recent patterns. Boundary: not for code review, not for PR review (use pr-review-toolkit).

Handle cspell unknown word warnings with a prioritized decision tree and config bootstrapping. Use when users encounter cspell diagnostics, spelling errors from cspell, CI or linting failures on unrecognized words, or ask to add words to the cspell dictionary, suppress cspell warnings, or choose between cspell:words and cspell:ignore directives. Also trigger when setting up cspell in a new project, when any cspell-related IDE warning appears, or when a pre-commit hook fails due to unknown words. Boundary: not for non-cspell spell checkers (typo, codespell, Vale, textlint) or other linting tools (ESLint, markdownlint).

Manage agent skills using the npx skills CLI. Use when the user asks to "install a skill", "add a skill", "remove a skill", "list skills", "update skills", "find skills", "search for skills", or mentions "npx skills", skill management, or agent skill installation. Also trigger when the user mentions skills.sh, agent-skills, vercel skills, or wants to discover available skills for their environment. Boundary: not for creating new skills (use skill-creator) or managing plugins (/plugin commands).

Generate a meaningful session name based on the overall session theme, not just the last action. Use when asked to "name this session", "rename session", "prename", "give this session a name", "what should I call this session", or when the user invokes /prename. Outputs a concise session title, with a platform-specific rename command only when the current agent supports one. Boundary: not for renaming files, variables, or git branches.

Replace one library with another (e.g., moment.js to date-fns, webpack to vite), or migrate deprecated API patterns within the same library (e.g., React 19 forwardRef removal, Vue 3 Options API to Composition API). Use when asked to "replace X with Y", "migrate from X to Y", "switch from X to Y", "swap X for Y", "convert from X to Y", "port from X to Y", "remove deprecated forwardRef", "migrate to new API", or when planning library replacement or API migration. Also trigger when the user mentions switching libraries, finding alternatives, or removing deprecated patterns. Boundary: for replacing libraries or migrating API patterns. Use deps-upgrade for version bumps within the same library.

Upgrade dependencies with breaking change detection and migration planning. Use when asked to "upgrade dependencies", "update packages", "bump dependencies", "bump <package> to <version>", "update <package> to version <version>", "handle dependabot PR", "handle renovate PR", "check for breaking changes", "upgrade <package> to <version>", "check outdated dependencies", "scan for updates", or when reviewing dependency bot PRs (dependabot, renovate, GitHub Actions), or when addressing security alerts, CVE notices, or vulnerability patches. Also trigger when the user mentions outdated packages, version bumps, or package manager update commands. Boundary: for version bumps within the same library. Use deps-migrate when replacing one library with another.

Guided focus questioning to align on requirements before planning or complex tasks. Use when asked to "focus first", "ask me questions before planning", "guided focus", "help me clarify requirements", "let's align before starting", or proactively when detecting complex/ambiguous tasks that would benefit from requirement clarification before entering plan mode or execution. Always ask the user whether to activate before starting proactively. Boundary: not for simple, unambiguous tasks with clear requirements.

Parse and verify checklists from GitHub PRs and issues, auto-checking items that pass verification. Use when asked to verify PR/issue checklists, check off items, process checklist tasks, run the checklist, or when a PR/issue contains unchecked checkbox items that need verification. Also trigger when the user mentions "checklist", "checkbox", or wants to auto-verify PR merge requirements. Boundary: not for creating new checklists (just write markdown) or handling PR review comments (use pr-comment-resolver).

Build and merge .gitignore files using github/gitignore templates with smart project detection. Use when the user asks to "create gitignore", "build .gitignore", "add gitignore templates", "set up gitignore", "update gitignore", or requests "/gitignore". Also trigger when observing projects without .gitignore, seeing untracked files like node_modules/, .env, __pycache__/, or *.log in git status, or after git init in a new repo. Boundary: not for .dockerignore or other non-git ignore files.

Rebase stacked PRs after parent PR is merged, preserving only your commits. Use when the user asks to "rebase my PR after parent merged", "update stacked PR", "fix PR after dependency merged", "cherry-pick my commits to new base", "sync stacked PR", or when a PR contains commits from a merged parent PR that need to be removed. Also trigger when the user mentions stacked PRs, dependent PRs, PR chains, or parent PR merges affecting their branch. Boundary: not for regular rebases onto main or resolving merge conflicts in non-stacked PRs.

Automate PR comment review, fix, and resolution workflow with atomic commits. Use when the user asks to "handle PR comments", "resolve PR review comments", "fix PR feedback", "process review comments", "address PR suggestions", "deal with review comments", or provides a GitHub PR URL with review comments. Also trigger when the user mentions unresolved PR threads or wants to batch-process reviewer feedback. Boundary: not for writing PR reviews (use code-review) or PR checklists (use checklist-runner).

Cleanly uninstall applications on macOS with thorough research and cleanup. Use when the user asks to "uninstall", "remove", "delete", or "clean up" an application, program, CLI tool, or package on macOS. Also trigger when the user wants to check what residual data an app has left behind, asks to "check leftover files", or mentions cleaning up after an app removal. Boundary: macOS only. Not for Linux/Windows, removing SIP-protected system apps, or clearing browser data.

TypeScript coding guidelines with dos and don'ts for type design and patterns. Use when writing, reviewing, or refactoring TypeScript code in projects with tsconfig.json or .ts/.tsx files. Trigger when the user asks to "review TypeScript code", "check my TS code", "write TypeScript", or when creating or modifying any .ts/.tsx file. Also applies when discussing type design, generics, naming conventions, interface vs type decisions, or TypeScript patterns. Boundary: not for JavaScript-only projects without TypeScript configuration.