llm

v0.1.0

Multi-LLM research — deep research, second opinions, and multi-model debate via GPT, Gemini, Grok, and Perplexity

Claude Code5 Skills

By beornLicense: MIT3 GitHub starsUpdated 1 hour ago

Directory evidence

Runtimes: Claude Code
Parsed components: 5 skill or MCP entries
Source updated: Sep 26, 2026
Manifest status: Canonical path parsed

The directory validates manifest shape and source location. It does not execute the plugin or provide a security endorsement. Review the indexing methodology →

Install llm for Claude Code

Installs for the current user

claude plugin marketplace add IchenDEV/agent-plugin-mkt
claude plugin marketplace update agent-plugin-marketplace
claude plugin install llm@agent-plugin-marketplace

Paste and run these commands in a terminal with Claude Code. They add and refresh the PluginsMP catalog, then install this plugin.

The installer fetches third-party code from the source repository shown on this page. This directory validates manifest structure and source location, but does not perform a security audit; review the manifest, components, and source before installing.

Get the source manually

$ git clone https://github.com/beorn/bearly

Clone the source repository, then follow its setup instructions to add the plugin to a compatible client. The plugin root is plugins/llm/.

Plugin files

plugins/llm/

├── .claude-plugin/plugin.json
├── skills/ask/SKILL.md
├── skills/big/SKILL.md
├── skills/deep/SKILL.md
├── skills/fresh/SKILL.md
└── skills/llm/SKILL.md

Included Skills5

askskills/ask/SKILL.md

Single-model quick questions to other LLMs — fast, cheap (~$0.02). Use for one-off second opinions and prior-art lookups. For multi-model judging use /pro. For web-search research use /deep.

bigskills/big/SKILL.md

META-PROTOCOL for reframing the problem (10-20 hypotheses, 2 rounds, find the design where the bug can't happen). Calls /pro or /deep internally — not itself an LLM tool. Use when the fix feels like a patch or the same area keeps breaking. Subsumes /fresh.

deepskills/deep/SKILL.md

Long-running web-search research with citations via OpenAI's Deep Research API (~$2-5, 2-15 min, fire-and-forget). Use when prior art / external citations matter. NOT DeepSeek. For multi-model judging without web search use /pro.

freshskills/fresh/SKILL.md

META-PROTOCOL for being stuck 20+ min on a specific problem — stops coding, gathers context, calls /deep (or /pro) internally with a structured request. Not itself an LLM tool. For unstructured stuck-feelings use /big; for direct questions use /ask, /pro, /deep.

llmskills/llm/SKILL.md

Multi-LLM research — deep research, second opinions, and multi-model debate. Use when user says /llm.

Plugin manifests1

plugins/llm/.claude-plugin/plugin.json

{
  "name": "llm",
  "version": "0.1.0",
  "description": "Multi-LLM research — deep research, second opinions, and multi-model debate via GPT, Gemini, Grok, and Perplexity",
  "author": {
    "name": "beorn",
    "url": "https://github.com/beorn"
  },
  "repository": "https://github.com/beorn/bearly",
  "license": "MIT",
  "keywords": [
    "llm",
    "research",
    "gpt",
    "gemini",
    "grok",
    "multi-model",
    "debate"
  ]
}