Skills
23 skills · 100% produce the artifact · 0% audits · none measured yet

A Claude skills marketplace. 23 skills. Every one makes something.

Search them, install them with one command, no account. Each one produces a finished piece of work, not a report about your work. How we test →

One command

Add the marketplace, then take what you need.

This directory is a Claude Code plugin marketplace as well as a web page. Adding it costs one command, and it needs no account, no desktop app and no working out which folder a skill belongs in.

/plugin marketplace add mkhalid1/locul-skills
  1. Add the marketplace. Open Claude Code and run the command above. It registers this directory as a plugin marketplace. Nothing is installed yet, and nothing is sent anywhere.
  2. Install the bundles you want. Run /plugin to browse the bundles, one per task category, and install the ones you want. Claude Code puts the skills where it already looks for them.
  3. Or take the file on its own. Install nothing at all, if you would rather not. Every skill page carries its raw SKILL.md to copy or download. These files are plain markdown with no tool bindings, so they also work in Claude Desktop, Cursor, and anything else that reads an instruction file.

The marketplace is a public GitHub repository, which is the whole distribution: github.com/mkhalid1/locul-skills. You can read every file before you install it, and updates arrive when you pull.

All 23 shown

23 skills we have not measured yet

Written, reviewed and free to take. No run behind them, so no claim about what they do to an output. What that means.

Paid adsNot yet measured Ad to landing page message match rewriter Takes the live ad and the current page copy and returns the rewritten page section by section, matched on six named points, with the branch rule for when one page cannot serve every ad. skill · 3,196 words · MIT Read the write-up
Resumes and job searchNot yet measured ATS resume tailorer Separates the knockout questions that actually reject applications from the match score that only sorts them, then rewrites the CV against what parsers and recruiter search are documented to do. skill · 3,334 words · MIT Read the write-up
Growth and launchNot yet measured Email list giveaway campaign builder Prize-first, coefficient-first: the arithmetic that predicts how many of a giveaway's entrants become real subscribers, before a partner is contacted or money is spent. skill · 3,391 words · MIT Read the write-up
Paid adsNot yet measured Facebook ad set writer Matches objective to funnel stage, sizes budget and audience against real floors, writes three ad variants pinned to distinct awareness levels, and sets the numeric 72-hour kill or scale rule. skill · 3,968 words · MIT Read the write-up
Paid adsNot yet measured Google Ads negative keyword list builder Emits an account list, campaign lists and ad group lists, every entry carrying an explicit match type, built around the documented fact that negatives do not match close variants. skill · 3,082 words · MIT Read the write-up
Paid adsNot yet measured Google Search RSA writer Assigns every headline slot a distinct job before writing a word, prices each pin in lost combinations, and ends on an impression floor that says when you have learned nothing yet. skill · 3,293 words · MIT Read the write-up
SEONot yet measured Internal link map builder Produces the actual link map, direction by direction, with anchor text held to Google's documented standard and changes sequenced into dated batches you can verify one at a time. skill · 3,248 words · MIT Read the write-up
SEONot yet measured Keyword cluster map builder Clusters keywords by SERP result overlap rather than semantic similarity, then rules on which pages to build, merge or drop. skill · 3,256 words · MIT Read the write-up
LinkedInNot yet measured LinkedIn carousel builder Settles which of the three formats named carousel you actually want, plans the pages, and runs the verification pass that a frozen-after-posting file makes compulsory. skill · 3,529 words · MIT Read the write-up
LinkedInNot yet measured LinkedIn comment writer Ranks the room before ranking the words: an audience-overlap test decides whether to comment, six archetypes earn Most Relevant, and a decision rule closes with a private-message branch. skill · 3,615 words · MIT Read the write-up
LinkedInNot yet measured LinkedIn DM sequence writer Rations a free account's small monthly note allowance across a target list, sets the trigger that justifies each approach, and writes a three-message sequence whose first message carries no ask. skill · 3,571 words · MIT Read the write-up
LinkedInNot yet measured LinkedIn founder profile rewriter Picks one audience, then rewrites headline, About, Featured, Experience, Skills, cover and primary action against LinkedIn's documented profile constraints. skill · 3,653 words · MIT Read the write-up
LinkedInNot yet measured LinkedIn post generator Picks the LinkedIn post format from raw material, writes a hook that survives an undocumented fold, and treats hashtags, the link question and the golden hour exactly as documented. skill · 3,788 words · MIT Read the write-up
Paid adsNot yet measured Meta ads copy generator Builds a pastable Meta copy set: three variants pinned to awareness levels, written to the fields each placement really renders, and checked against the personal attributes policy before launch. skill · 3,133 words · MIT Read the write-up
Paid adsNot yet measured Performance Max asset group builder Emits a build-ready Performance Max asset group: every text asset, image and video slot, search themes, audience signal and exclusions, with the caps Google actually publishes. skill · 3,139 words · MIT Read the write-up
Resumes and job searchNot yet measured Recruiter and hiring manager outreach writer Builds the exact search surface that makes you findable, then rations a free account's few 200-character notes across the targets whose trigger actually justifies an approach. skill · 3,207 words · MIT Read the write-up
Resumes and job searchNot yet measured Role-matched cover letter writer A six-branch rule for whether a letter is worth writing at all, then one claim attached to evidence a reference check could confirm, inside a stated word budget. skill · 3,195 words · MIT Read the write-up
Growth and launchNot yet measured SaaS launch promotion plan builder Scores SaaS launch channels on reach, fit, proof and cost, sequences the checklist around a working trial and staffed support, and measures yield against trial-to-paid conversion and payback. skill · 3,382 words · MIT Read the write-up
Resumes and job searchNot yet measured Salary negotiation script writer Builds the script around who is legally obliged to name a number first and who is barred from asking your pay history, then scripts the exchanges that follow from that. skill · 3,197 words · MIT Read the write-up
SEONot yet measured SERP-first SEO article writer Classifies what a keyword's results page rewards, matches an article's format, depth and structure to it, then checks the draft against the three real competitors before it can pass. skill · 3,183 words · MIT Read the write-up
SEONot yet measured SERP title and meta description writer Replaces the unsourced 50-to-60-character rule with Google's documented title rewrite triggers, a width check needing no invented constant, and a branch for when a description is worth writing. skill · 3,120 words · MIT Read the write-up
Resumes and job searchNot yet measured STAR interview story bank builder Turns a career history into 8 to 12 indexed stories, each mapped to every competency it can serve, plus a coverage table naming the competencies with nothing behind them. skill · 3,199 words · MIT Read the write-up
SEONot yet measured Traffic decay content refresher Separates a real ranking decline from a Search Console measurement artefact or a retired rich result, then rewrites, consolidates, leaves alone or removes the page, with a record of why. skill · 3,330 words · MIT Read the write-up
By task

Browse by task

Skills are organised by the job you are doing, not the job title you have.

LinkedIn Posts, profiles, comments, carousels and outreach, written against character limits and truncation points rather than against advice. 5 skills · 0 measured SEO Search work built from primary sources: what the engines actually document, what they explicitly tell you to ignore, and the numbers everyone publishes that have no source at all. 5 skills · 0 measured Resumes and job search Applications, interviews and offers, built on what applicant tracking vendors and employment regulators actually publish, which contradicts most of the advice in circulation. 5 skills · 0 measured Paid ads Campaign work where the hard part is not the copy. It is the floors that decide whether a result means anything, the published caps and match rules nothing in the interface enforces for you, and the difference between a documented number and a repeated one. 6 skills · 0 measured Growth and launch Deciding what to promote through which channel, in what order, and how to tell afterwards which channel actually deserves next quarter's budget. 2 skills · 0 measured
The evidence

What the cards claim, and what they do not

They produce the artifact. 19 of these build something that did not exist before, and 4 rewrites something you already have. That is 100 per cent of the library, against 0 per cent that audit. The distinction is the whole filing rule and it is computed from a table in the repository rather than typed by hand, so it cannot quietly drift.

This is a correction, and it is worth being blunt about what it corrects. An earlier version of this directory ran to over a hundred skills and roughly four out of five of them were audits: files that read something you had already made and handed back a list of findings. Several were good. Almost none was a thing a person would go looking for, because an audit is usually a one-off errand rather than a job you do again on Tuesday. That library was withdrawn in full in August 2026 rather than pruned, and this one was built from an empty directory.

The platform facts are dated, or they are not here. Most published advice about any platform is folklore with a confident number attached. Where a constraint is documented, these files state it with the date it was checked and a route back to the source. Where the number in circulation turns out to be unverifiable, and a surprising amount of it is, the file says so and the method is built so it does not depend on the answer. A skill that repeats a plausible figure nobody can confirm is worse than one that admits the gap, because you cannot tell which half to trust.

None of them is measured, and every card says so. There is a harness behind this project. It runs a realistic task on real material, runs the identical task again with the skill removed, 5 runs each way, and grades every output alone against a rubric written by someone who has never seen the skill, by a session that is not told the other arm exists. It has been used. It has deleted skills that failed, including one that was statistically significant and still cut because the size of the effect came in under a line fixed before the first run. Nothing in this library has been through it yet. Until something has, the cards claim nothing about what these files do to an output.

Unmeasured is not a synonym for untrustworthy, and it is not a synonym for tested either. It means we have not run the experiment, so we are not going to tell you the result. The full method, the historical results table, and the round we had to throw away.

Free to copy. MIT. Mac and Windows.
Questions

Claude skills marketplaces, answered plainly

Is there a marketplace for Claude skills?

Yes, and there is more than one, because Claude Code ships the mechanism rather than the shop. Anyone can publish a plugin marketplace: you add one with a single command and browse what is inside it. This page is one of them. Others index public GitHub repositories in bulk and run to hundreds of thousands of entries, which is a search problem rather than a library. This one holds 23 skills across 5 task categories, each written for a specific job and each labelled with what we have and have not tested.

Can I use Claude skills for free?

Yes. Every skill here is free, MIT licensed, and needs no account. There is no email to hand over and nothing to sign up for. The install command is free, the raw SKILL.md on every skill page is free to copy or download, and the full source of each skill is printed on its own page so you can read the whole thing before you take it. Locul, the desktop app this directory came out of, has paid tiers. Nothing on this page sits behind them.

Where to browse skills in Claude?

Inside Claude Code, run /plugin once you have added a marketplace: that lists the bundles it carries and installs the ones you pick. Claude Desktop manages the skills you have already added through its settings panel. To browse before you install anything, use a web directory like this one, where you can search 23 skills by name, task or trigger, read the full source, and see the section on each page that says what the skill does not do.

What does Claude Marketplace do?

A plugin marketplace is a git repository with a .claude-plugin/marketplace.json catalogue in it. Adding one tells Claude Code where to find a set of plugins. Installing a plugin copies its skills, and any commands or agents it ships with, into your setup, and updates arrive the next time you pull. That is the whole mechanism. It is distribution plumbing rather than a store: there is no payment, no ranking, and no approval queue in it, which is also why the labelling on whatever you install is worth reading.

How to setup Claude marketplace?

Three steps, and the first two are commands. Run /plugin marketplace add mkhalid1/locul-skills in Claude Code to register this directory. Run /plugin to see the bundles and install the ones you want. Start a new session, and Claude Code picks each skill up from the name and description in its frontmatter, so you can also call one by name. The same shape works for any marketplace: swap the owner and repository in the first command.

Adding skills on top of a CLAUDE.md you have not read in months is how a setup gets confusing. There is a free CLAUDE.md analyser on this site that reads yours back to you before you stack anything else on it, and it needs no account either.

Is there an official Anthropic skills marketplace?

No. Anthropic documents the plugin marketplace mechanism and publishes plugins and example skills of its own, but it does not run a storefront where third-party skills are listed, ranked or approved. Every marketplace you add is built by somebody else, including this one. Anything advertising itself as the official Anthropic skills marketplace is not one, and that is worth knowing before you install from it.

Is the marketplace on GitHub?

Yes, and the repository is the distribution rather than a mirror of it. The install command points at github.com/mkhalid1/locul-skills, which is public. You can read every file before installing it, diff an update before pulling it, and open an issue against a skill that is wrong. Nothing about this library needs the desktop app: the app is a second path for people who already have it, not a gate in front of the files.

One click, or no click at all

Import a skill straight into your setup.

Locul puts a skill in the right folder for Claude Code, Claude Desktop and Codex, scrubs any environment variables before they reach you, and keeps the memory those skills read from up to date by itself. Mac and Windows.

Start free