Browse Secure Claude Skills
Search verified agent skills and review security grades before installing · full A–Z index
- Playwright CliAutomate browser interactions, test web pages and work with Playwright tests.Votes: 0GitHub stars: 2
- Locomotive ScrollComprehensive skill for Locomotive Scroll smooth scrolling library with parallax effects, viewport detection, and scroll-driven animations. Use this skill when implementing smooth scrolling experiences, creating parallax effects, building scroll-triggered animations, or developing immersive scrolling websites. Triggers on tasks involving Locomotive Scroll, smooth scrolling, parallax, scroll detection, scroll events, sticky elements, horizontal scrolling, or GSAP ScrollTrigger integration. Int...Votes: 0GitHub stars: 2
- T4 Dev WorkflowUse when planning or implementing a feature/change in a T4-team repo (T4 Labs / Slow-Inc) — deciding how to go from idea to shipped code, filing or updating a GitHub issue or PRD, writing a bilingual (Thai + English) issue/PR body, opening a PR, or closing an issue. Covers the grill→PRD→issues→TDD pipeline, the PRD→issues→PR gate, the auto-triggered skill map, triage labels, and the issue lifecycle. Triggers include "let's build X", "file an issue for this", "write the PRD", "open a PR", "wha...Votes: 0GitHub stars: 2
- Qwen38 ThinkThe thinking a small model skips, as steps it must write down: before the first edit, an assumptions table and a three-flaw check on the brief (contradiction, missing fact, impossible ask), each flaw checked by a command; when stuck, one observation, one hypothesis, one falsifying command; and a THINK report line. Pushback is a written line in the report, never a question that stops the work. Slice 2 of xeno-skills #355.Votes: 0GitHub stars: 2
- Qwen38 Claude CodeThe goal: Qwen3.8-27B using the Claude Code harness as well as it possibly can, guided rather than guessing. A verdict for every tool Claude Code 2.1.281 sends (use / use when X / do not, with the reason), which tool for which want, the harness facts no schema states (the 12,288-token turn cap, images through Read, screenshots, what Bash blocks), what to do when a tool returns an error including a harness denial, and the CC report line. The family's one full-coverage skill (#358, #380).Votes: 0GitHub stars: 2
- Using ClinkThe entry point for delegating to another AI CLI agent through PAL's clink tool. Decides which of the four clink skills you actually want before you spend a call — executing a scoped subtask, convening a panel for judgment, hunting a bug across agents, or deciding who may do the work at all. Reach for it whenever you are about to delegate, whenever you are choosing between a subagent and a panel, and whenever a delegation came back wrong and you are not sure the routing was right. Triggers in...Votes: 0GitHub stars: 2
- Clink MasteragentWhat a master agent owns and can never delegate, and how to pick the model for a delegation from measured scores instead of memory. Use before any clink delegation, and whenever choosing between models, efforts, or clients — including "which model should review this", "is the cheap one good enough", "should I escalate the effort". Carries the full per-axis score table for every reachable model+effort inline, generated from the structured source, so the choice is made against data rather than ...Votes: 0GitHub stars: 2
- Clink DebugThe single home for delegating a bug hunt to clink agents — read it before sending any failure to clink-subagents or clink-brainstorm, not only a bug that already survived a pass. It routes a small bug to one cheap worker and escalates only what survives, defines the evidence that must travel with it, and enforces that no agent which produced a hypothesis may falsify or repair it. Use when a bug is about to be delegated, is still open after a worker looked at it, has two agents disagreeing ab...Votes: 0GitHub stars: 2
- Design Ship GateDone-gate for a web page built by a small model. Run it before saying a landing page is finished: brief coverage, then nine executable checks, one per defect seen in Qwen3.8-27B pages (fonts, OG tags, dark mode, 390 px overflow, number wrapping, hero collisions, tracking, language, a script that does not parse). Commands and pass conditions, no design theory.Votes: 0GitHub stars: 2
- Design RulesUniversal micro-UI/UX design & engineering rules based on Kole Jain, Chris McCoy, Ran Segall, and Satori Graphics' frameworks. Enforces Pro Typography hacks (-2% tracking, Major Third 1.25x scale, 150% body line-height), 60-30-10 color balance, Material opacity modulation, 12/8/4 column grids, 8pt spacing rhythm, 4-state buttons, scrolling CTA frequency, progressive backdrop blur overlays, Luxury White Space, and the LIFT System with 6 Levels of Visual Flow.Votes: 0GitHub stars: 2
- Ask XenoThe top of this library. Ask it which family your task belongs to, then take the family's own route from there. Reach for it when you are unsure which skill applies, when a task spans families, or when you suspect a skill exists for this and cannot name it. Deliberately thin — it names the family entries and nothing else, so it stays cheap enough to carry everywhere. Triggers include "is there a skill for X", "which skill should I use", "how do we do X here", and any moment you are about to i...Votes: 0GitHub stars: 2
- Ship ItEnd-to-end ship a change or coordinate an approved complex plan — implement, two-pass adversarial code review via the adversarial-review skill, adversarial manual testing via the adversarial-test skill, open PR, wait for Copilot review + green CI, address feedback, merge. Accepts a plain task description (no issue is created unless --issue is passed), a GitHub issue URL (closed on merge), or a Jira ticket URL (read-only; the key goes in the PR). Use when asked to implement and ship a change e...Votes: 0GitHub stars: 5
- Youtube SummarySummarize a YouTube video from its transcript into a concise Markdown note with a TL;DR, key points and timestamps, then save it to the notes folder set by YOUTUBE_SUMMARY_DIR (for example an Obsidian inbox) or print it in chat. Use when the user shares a YouTube URL and wants a summary, key takeaways, or notes from a video.Votes: 0GitHub stars: 5
- Adversarial TestAdversarial manual testing of a change. A clean-context Test Adversary subagent derives acceptance criteria from the task, runs the real product surface (service, CLI, sandbox) in isolated state, and attacks boundaries, malformed input, repetition, and adjacent flows; every reproduction is then rerun to drop hallucinated failures. Use to prove a branch or PR actually works before merging, or as the testing gate inside ship-it.Votes: 0GitHub stars: 5
- Lazygitlazygit is a keyboard-driven terminal interface for Git that makes staging single lines, interactive rebase, cherry-picking, bisecting and undoing mistakes a matter of a few keypresses. Use when a user asks to install lazygit, configure its config.yml, add custom commands or change keybindings, set up delta as its diff renderer, or wants to know which keys to press to squash, amend, cherry-pick or undo commits in lazygit.Votes: 0GitHub stars: 155
- ShortcutsManage and interact with macOS Shortcuts. Use this skill when the user wants to list available shortcuts, view a shortcut in the Shortcuts app, or run a shortcut. Supports running shortcuts with or without input parameters.Votes: 0GitHub stars: 3
- Adversarial ReviewFast two-pass adversarial code review. A Code Adversary subagent red-teams a diff or PR to find the concrete bug, then a separate Findings Adversary subagent with a clean context tries to refute every finding to cut false positives. Use for routine changes where a full staff review is overkill, or as the review gate inside ship-it.Votes: 0GitHub stars: 5
- Windags DecomposerFirst-party three-pass planning protocol that converts a ProblemUnderstanding into a revisioned decomposition proposal: typed nodes, reasoned edges, uncertainty records, candidate skills, and topological waves. A DAG is not an execution proof; confidence values are not probabilities without calibration. NOT for execution, runtime admission, or formal HTN proofs.Votes: 0GitHub stars: 2
- Task DecomposerTurn a stated outcome into a reviewable task graph with deliverables, decision/unknown nodes, typed dependency reasons, and authority gates. It plans work; it does not execute, schedule, or prove that a DAG can run concurrently. NOT for runtime orchestration or granting effect authority.Votes: 0GitHub stars: 2
- Rust Development WorkflowRun the ordinary Rust edit, compile, lint, and test loop; diagnose ownership and lifetime errors with intent and surrounding code; choose basic error and async test patterns. Use for day-to-day Rust changes in any crate. For library API invariants use advanced-rust-patterns; for structure selection use rust-data-structures-advanced; for measured optimization use rust-performance-and-idioms; for runtime failures use rust-debugging-mastery; for a pd-console pane or GPUI executor boundary use gp...Votes: 0GitHub stars: 2