# emdly

> emdly is a catalog of Markdown skills for AI agents: plain .md playbooks (instructions, rubrics, output contracts) that you browse, upvote and install into Claude Code with `npx @emdly/cli add owner/skill`, fetch as raw Markdown at a stable URL, or load over MCP. Every published skill passed automated checks (Markdown lint, license, prompt-injection scan), an AI safety scan and human review, and is versioned. Next to the catalog, every member has a private Brain: named sets of Markdown memory files (what is true in their context) that an agent reads and writes over MCP, encrypted at rest and versioned.

Skills are identified as `owner/skill-name`. Raw Markdown: `https://emdly.com/raw/{owner}/{skill}.md` (latest) or `…@v12.md` (pinned). Any HTML page below can be fetched as Markdown by adding `.md` to its path or sending `Accept: text/markdown`.

## Docs
- [Quickstart, raw endpoints, REST API, MCP server](https://emdly.com/docs.md): how to install and integrate
- [Publishing & review](https://emdly.com/docs.md#publishing--review): the pipeline every submission goes through
- [Agent skills — format, install, versions](https://emdly.com/docs/skills.md): what a skill is and how to use one
- [Brain — private agent memory](https://emdly.com/docs/brain.md): named brains, sharing, MCP tools
- [Badges — Verified, Recommended, Certified](https://emdly.com/docs/badges.md): what the marks on authors and skills mean

## Catalog
- [All skills](https://emdly.com/llms-full.txt): every published skill with description and raw URL
- [Code review](https://emdly.com/index.md?cat=code-review): 4 skills
- [Copywriting](https://emdly.com/index.md?cat=copywriting): 5 skills
- [Data & analytics](https://emdly.com/index.md?cat=data-analytics): 8 skills
- [Design](https://emdly.com/index.md?cat=design): 1 skill
- [Development](https://emdly.com/index.md?cat=development): 15 skills
- [Docs & changelogs](https://emdly.com/index.md?cat=docs-changelogs): 3 skills
- [Ecommerce](https://emdly.com/index.md?cat=ecommerce): 10 skills
- [Email & outreach](https://emdly.com/index.md?cat=email-outreach): 5 skills
- [Finance](https://emdly.com/index.md?cat=finance): 1 skill
- [Game development](https://emdly.com/index.md?cat=game-development): 2 skills
- [Hiring](https://emdly.com/index.md?cat=hiring): 1 skill
- [Learning](https://emdly.com/index.md?cat=learning): 1 skill
- [Marketing](https://emdly.com/index.md?cat=marketing): 3 skills
- [Product](https://emdly.com/index.md?cat=product): 1 skill
- [Productivity](https://emdly.com/index.md?cat=productivity): 3 skills
- [Research](https://emdly.com/index.md?cat=research): 3 skills
- [Security](https://emdly.com/index.md?cat=security): 2 skills
- [SEO](https://emdly.com/index.md?cat=seo): 3 skills
- [Support](https://emdly.com/index.md?cat=support): 1 skill
- [Team ops](https://emdly.com/index.md?cat=team-ops): 2 skills
- [Ticket ops](https://emdly.com/index.md?cat=ticket-ops): 3 skills
- [Utilities](https://emdly.com/index.md?cat=utilities): 1 skill

## Most upvoted
- [querydeck/sql-query-explainer](https://emdly.com/raw/querydeck/sql-query-explainer.md): Explains what a query actually does — joins, filters, gotchas — in plain language, then flags the index it wishes existed.
- [kernelpanic/pr-review-ritual](https://emdly.com/raw/kernelpanic/pr-review-ritual.md): A reviewing discipline for agents — read the diff twice, test the edge cases, comment on intent, never nitpick formatting a linter owns.
- [shopmetric/ga4-funnel-analyst](https://emdly.com/raw/shopmetric/ga4-funnel-analyst.md): Reads a GA4 export and finds where the checkout funnel leaks — drop-off by step, device and source — with the one fix to try first.
- [opsmith/jira-ticket-scorer](https://emdly.com/raw/opsmith/jira-ticket-scorer.md): Scores every Jira ticket for clarity, scope and effort before it hits the sprint. Flags vague acceptance criteria and suggests a rewrite.
- [pixelforge/playtest-feedback-triage](https://emdly.com/raw/pixelforge/playtest-feedback-triage.md): Sorts raw playtest notes into balance, UX, bugs and feel — ranked by how many testers hit it, with repro steps pulled from context.
- [sevzero/incident-postmortem](https://emdly.com/raw/sevzero/incident-postmortem.md): Drafts blameless postmortems from a timeline and a channel export — impact, contributing factors, and actions with owners.
- [fieldnotes/user-interview-distiller](https://emdly.com/raw/fieldnotes/user-interview-distiller.md): Distills interview transcripts into claims with evidence — what was said, how often, and what it contradicts in your assumptions.
- [outboundlab/cold-email-first-touch](https://emdly.com/raw/outboundlab/cold-email-first-touch.md): Writes first-touch e-mails that sound like a person — one observation, one sentence of value, one soft ask. Kills every buzzword on sight.
- [tonecheck/brand-voice-guard](https://emdly.com/raw/tonecheck/brand-voice-guard.md): Holds any draft against your voice guide — banned words, sentence rhythm, claims policy — and returns a marked-up pass, not a rewrite.
- [opsmith/standup-synthesizer](https://emdly.com/raw/opsmith/standup-synthesizer.md): Reads yesterday's commits, tickets and threads, and writes each person's standup draft — blockers first, no ceremony.

## Blog
- [emdly is live — a home for the files your agents read](https://emdly.com/blog/emdly-is-live.md): Today we open emdly to everyone. A catalog of agent skills in plain Markdown, reviewed before they go live, installable with one command or straight from Claude over MCP. Here is what is in the box and what comes next.
- [Connect Claude to emdly: skills and your Brain over MCP](https://emdly.com/blog/connect-claude-to-emdly.md): One URL turns the whole catalog into tools Claude can call, and a browser sign-in gives it your private Brain. Setup for Claude Desktop, claude.ai and Claude Code, and what each tool does.
- [Why agent skills should be plain Markdown](https://emdly.com/blog/why-agent-skills-should-be-plain-markdown.md): Prompts scattered across repos, docs and heads. A skill is just a .md with a contract — here is why that beats YAML, JSON and databases.
- [Writing a skill that survives review](https://emdly.com/blog/writing-a-skill-that-survives-review.md): A skill is not a prompt. The structure that makes one reusable, the four sections reviewers look for, and a template you can publish in ten minutes.
- [How review works: what we approve and what we bounce](https://emdly.com/blog/how-review-works.md): Every submission runs lint, license and prompt-injection checks and an AI safety scan before a human reads it. The most common rejection reasons, with examples.

## Optional
- [REST API](https://emdly.com/v1/skills): JSON, bearer token for writes
- [MCP server](https://emdly.com/mcp): tools search_skills, get_skill, install_skill
- [Leaderboard](https://emdly.com/leaderboard.md)
- [Contributors](https://emdly.com/contributors.md): every author with a published skill