Sync vendored third-party skills against their upstream repositories, re-apply local patches, and flag new upstream skills. Invoke explicitly with /sync-skills.
This skill should be used when creating a GitHub pull request for the Composio SDK repository. It generates conventional commit-style PR titles and structured descriptions with Linear issue links, concise bullet points, and optional mermaid diagrams. Triggers on "create PR", "open PR", "make a pull request", or when the user is ready to submit changes for review.
Commit completed units of work as you go instead of batching everything into one commit at the end. Use during any coding session that will produce more than one logical change: after each green test run, when a logical unit is complete, or before starting a risky refactor. Also applies when the user says "commit often", "checkpoint", or "commit as you go".
Help users operate the published Composio CLI to find the right tool, connect accounts, inspect schemas, execute tools, subscribe to trigger events with `composio listen`, script workflows with `composio run`, and call authenticated app APIs with `composio proxy`. Use when the user asks how to do something with `composio`, wants to run a known tool slug, needs to discover a slug with `composio search`, fix a missing connection with `composio link`, inspect tool inputs with `--get-schema` or `--dry-run`, troubleshoot top-level CLI flows, or explicitly needs `composio dev` guidance.
Use when the user wants to design, redesign, shape, critique, audit, polish, clarify, distill, harden, optimize, adapt, animate, colorize, extract, or otherwise improve a frontend interface. Covers websites, landing pages, dashboards, product UI, app shells, components, forms, settings, onboarding, and empty states. Handles UX review, visual hierarchy, information architecture, cognitive load, accessibility, performance, responsive behavior, theming, anti-patterns, typography, fonts, spacing, layout, alignment, color, motion, micro-interactions, UX copy, error states, edge cases, i18n, and reusable design systems or tokens. Also use for bland designs that need to become bolder or more delightful, loud designs that should become quieter, live browser iteration on UI elements, or ambitious visual effects that should feel technically extraordinary. Not for backend-only or non-UI tasks.
Phase-driven workflow for building a feature end to end. Four phases run in order: spec gathers intent and constraints, plan turns the spec into commit-sized units, work executes the plan one commit at a time, and safe-change handles scoped edits where a full spec/plan cycle is overkill. Use when starting new work, when an existing plan needs to be executed, or when making a single change that still needs verification.
Invoke jk, @jkomyno's deterministic personal CLI, for mechanical agent-friendly tasks: read the authenticated GitHub user (whoami), send Telegram notifications or wrap a long command and notify on its exit status (notify), pin and check GitHub Actions refs (actions bump/check), and acquire / index / search per-idea reference corpora under ~/.jk/ideas (idea). Use this skill when running any `jk ...` command, deciding which jk command fits a task, or parsing jk output in a script or hook. It documents the output contract (piped stdout is JSON, a TTY is text, --json forces either) and the exit codes (0 ok, 1 findings, 2 usage, 3 operational).
Review diffs, PRs, plans, or specs for correctness, architecture drift, missing tests, and scope creep. Use when changes are about to ship, when a plan needs sanity-checking before execution, or when a spec needs to be stress-tested. Returns a structured findings report grouped by severity.