Agent Skills

Agent skills for shipping software

27 skills your coding agent loads when a task needs one, from UI audits and typography to PR review and releases.

npx skills add mblode/agent-skills

How it works

No prompt packs and no 500-skill dump: 27 workflows your agent follows step by step.

  1. Install once

    One command copies all 27 skills into the folder your agent reads.

  2. Ask as usual

    Ask for a UI audit, a PR review or a release the way you already do. No special syntax.

  3. The agent loads the match

    Your agent reads each skill's one-line description, opens the SKILL.md that fits, and follows it. The rest stay closed.

Installs and stars

Installs on skills.sh
26.9K
GitHub stars
127

All 27 skills

Architecture

Design

  • product-design

    What the interface should do, before anyone builds it.

  • ui-design

    Visual direction, design-system extraction, Tailwind builds, screenshot to markup, dark mode, responsive, and a React and Next.js UX audit with a ship verdict.

  • ui-verification

    Boots the app in a headless browser and runs nine probes that turn inferred UI defects into measured ones, then proves the fix cleared.

  • ui-animation

    Springs, gestures, easing, and curves pulled from a screen recording.

  • presentation-creator

    Decks with a story spine, speaker notes, and a contrast-checked QA pass, as Marp markdown, a web app, or a handoff to pptx.

Writing

  • eli5

    Session talk in plain language: optional analogy, exact technical terms, and house vocabulary without forced templates.

  • ghostwriter

    Writes anything as you or as your company: messages, posts, PRs, tickets, PRDs, design docs, slides, copy, docs, READMEs. Works with no profile and writes one from pasted samples.

Quality

  • ax-audit

    Agentic experience audit: 27 rules for tool parity, approval gates, and escape hatches, ship verdict.

  • dx-audit

    Libraries, CLIs, SDKs, npm packages: 38 rules, agent-friendly checks, root-cause findings.

  • agent-ready

    Implements AFDocs, Is Agentic, Is It Agent Ready, and agent 404 findings: llms.txt maps, markdown twins, API errors.

  • typography-audit

    78 rules: punctuation, fonts, sizing, spacing, hierarchy, pairing.

  • seo

    SEO/AEO audits and fixes, search-demand research, writer briefs, and search, AI visibility, and conversion measurement.

  • test-audit

    Prunes low-value tests to a measured target with coverage held, and gates new tests before they land.

Shipping

  • planning

    Writes and reviews executable plans with repository evidence, vertical slices, and explicit acceptance criteria.

  • tidy

    Diff or PR review with file:line findings in confirmed and plausible tiers (bugs, structure, AI slop, security), report-only by default; apply mode lands the fixes and simplifies the diff.

  • pr-creator

    PRs with short human descriptions, Linear IDs, templates, drafts, tidied commits.

  • pr-babysitter

    Watches an open PR: conflicts, CI, comments. Fixes what it can.

  • autoship

    npm releases with changesets: fix loop, CI watch, Version Packages merge, OIDC publish verify.

  • ci-speedup

    Measures a pipeline's critical path as median and p90 over runs, then splits, shards, trims setup, and shares test module state, with a ledger and trend.

Authoring

  • chat-history

    Recover decisions, previous fixes, and context from past AI conversations with fast local search and source evidence.

  • agents-md

    Wires a repo so Claude Code, Codex and Cursor read the same instructions, migrates to native AGENTS.md without Claude wrappers, then audits and refactors instructions with grades.

  • agent-skills-creator

    Creates and audits skills by their added value, with portable workflows, a validator, and regression scenarios.

  • save-md

    Writes a named source as a markdown file the next turn can reread.

Questions

What is an agent skill?

A skill is a folder with a SKILL.md file: a name, a one-line description, and the workflow your coding agent follows. The agent reads every description up front and opens the full file only when a task matches, so an installed skill costs one line of context until it is needed.

Which coding agents do these skills work with?

Any agent that reads the Agent Skills format, including Claude Code and Codex. The installer asks which agents you use and puts the skills where each one looks for them.

How do I install a single skill?

Add --skill and the skill's name: npx skills add mblode/agent-skills --skill ui-design. Every skill on the board shows its own command above the workflow, so you can copy it from there.

Do I have to tell my agent to use a skill?

No. When a request matches a skill's description, the agent loads it on its own. You can still ask for one by name, for example "run the tidy skill on this diff", when you want a specific workflow.

How is this different from a prompt library?

A prompt is one message. A skill is a workflow: steps, rules and references the agent follows across a whole task, loaded only when the task calls for it. There are 27 here, each covering one job.

Are the skills free?

Yes. They are MIT licensed and the source is on GitHub. Read any SKILL.md before you install it, fork the repo, or copy one skill into your own set.

Where do the install and star numbers come from?

Installs are counted by skills.sh, which tracks installs made through the skills CLI. Stars come from the GitHub API. If either source cannot be reached, the number is hidden rather than guessed.

Install all 27 skills

Run it once. Each skill is there the next time a task needs it.

npx skills add mblode/agent-skills

Agent Skills