agentleFS
spscreamspscream/ai-skills/skills/trending-skills/SKILL.md

trending-skills

Skill1 starsChanged 3 months agoMITView on GitHub

Fetch fresh Claude Code / Cursor skills, agents, rules, MCP by topic with dedup.

  • Reads credentials
  • Installs packages

The file

We split this skill into the rules it gives an agent

2 rules. About 0 are common, 1 is distinctive, and 1 is very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈1

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 1

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Use when you need a weekly roundup of what's trending in the Claude Code / Cursor skill ecosystem: new skills, agents, rule files …

    line 18
  2. Distinctive:

    Search brings back more than it should: catalogue pages, marketing, articles that only mention the ecosystem in passing.

    line 73

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

trending-skills

Fetches fresh (last 7 days) trending skills / agents / rules / MCP for Claude Code and Cursor via Tavily search plus the GitHub API, dedups against previously-sent items, and outputs fresh items as a structured two-block digest for the LLM to format.

When to Use

Use when you need a weekly roundup of what's trending in the Claude Code / Cursor skill ecosystem: new skills, agents, rule files (.cursorrules / AGENTS.md), or MCP servers.

How to Run

python scripts/fetch_trending.py [--config config.yaml] [--dry-run] [--no-crawl]

--dry-run prints what a real run would emit but marks nothing as seen, so you can look at the output without consuming the items.

The script prints two blocks to stdout:

СТАТЬИ (заголовок | источник | url):            <-- fresh items
СТАТЬИ-СОДЕРЖАНИЕ (заголовок === текст):         <-- short crawled content per item

The agent (LLM) reads these blocks and formats the final human-readable post. Dedup is built in: a repeat run returns only genuinely new items, safe to call weekly.

Config

Everything is configurable via config.yaml (or the defaults at the top of the script). Sample:

queries:
  - {"q": "claude code skills trending", "domain": "github.com"}
  - {"q": "cursor rules file trending github", "domain": "github.com"}
  - {"q": "claude code MCP server new", "domain": null}
db_path: ./seen_skills.db
days: 7
max_results: 24
crawl_max_chars: 350
max_crawl: 6
tavily_api_key_env: TAVILY_API_KEY

# Second source: GitHub API (fresh repos by topic within `days`)
github_topics:
  - claude-code-skills
  - claude-code
  - cursor-rules
  - agents-md
  - mcp-server
github_max: 10
gh_token_env: GH_TOKEN      # optional: raises API limit 60 -> 5000 req/h

Requires: python -m pip install requests rapidfuzz pyyaml

What the editor must do with this output

Search brings back more than it should: catalogue pages, marketing, articles that only mention the ecosystem in passing. The collector cannot tell those from a real find, so filtering them out is your job.

  • Judge by meaning, not by the words in the title.
  • Drop anything that is not a skill, rule file, MCP server, agent or workflow for Claude Code or Cursor. Say nothing about what you removed.
  • A short roundup beats a roundup with a stray item in it. Fewer finds than you were aiming for is a normal outcome.

Output format (what the LLM turns into a post)

СТАТЬИ (заголовок | источник | url):
Title | github.com | https://...

СТАТЬИ-СОДЕРЖАНИЕ (заголовок === текст):
[Title]
<short text>
[КОНЕЦ]

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.