# llm-ai-agents-and-eng-research

llm-ai-agents-and-eng-research gathers recent updates on large language models, agent frameworks, and ML engineering while filtering out content older than seven days.

- subagent · `type=agent` · Claude Code, Codex, Cursor · ★ 3990 repo stars
- Author: disler
- License: not stated · policy=link-only · description only; the file stays with the author
- Source: https://github.com/disler/claude-code-hooks-mastery/tree/HEAD/.claude/agents/llm-ai-agents-and-eng-research.md
- Updated: 2026-03-04
- Security check: automated, 2026-11-16 (text) — no signs of malicious code found.
- HTML: https://affpapa.org/en/nevibecode/agents/llm-ai-agents-and-eng-research

## What it does

The subagent tracks recent model drops, benchmark evaluations, and multi-agent frameworks. It starts by running `date` in Bash to establish the time window. Everything older than seven days is discarded.

Research relies on `Bash`, `WebFetch`, and Firecrawl MCP tools (`mcp__firecrawl-mcp__firecrawl_search`, `mcp__firecrawl-mcp__firecrawl_scrape`). Without Firecrawl MCP configured in your environment, the subagent cannot crawl web sources or run external searches.

Findings are structured into distinct sections: key developments, model releases, autonomous agent tools, engineering practices, and practical takeaways. It highlights code repositories and performance benchmarks rather than speculative discussions.

## Install

### Claude Code (native platform)

**Manually: put the files in place**
- File: `.claude/agents/llm-ai-agents-and-eng-research.md` — For all projects use ~/.claude/agents/. Claude delegates by the description field or an @-mention

### Codex

**Manually: put the files in place**
- File: `.codex/agents/llm-ai-agents-and-eng-research.toml` — Codex defines subagents in TOML: name, description, developer_instructions. Move the Markdown text into developer_instructions by hand. For all projects use ~/.codex/agents/

### Cursor

**Manually: put the files in place**
- File: `.cursor/agents/llm-ai-agents-and-eng-research.md` — Cursor also reads .claude/agents/

Where a file is given instead of a command, copy the files from the source to that path: https://github.com/disler/claude-code-hooks-mastery/tree/HEAD/.claude/agents/llm-ai-agents-and-eng-research.md

## When to use it
- summarize language model releases and evaluations from the past seven days
- track updates across multi-agent frameworks and tooling repositories
- collect practical engineering techniques for deploying and optimizing ML systems

## Example prompts
- Run a research pass on new multi-agent frameworks released this week
- Summarize recent benchmark results and architecture updates for open weights models
- Find the latest engineering practices for LLM deployment and list usable libraries

## Similar
- [llm-council](https://affpapa.org/en/nevibecode/skills/llm-council.md) — skill · 1600 installs
- [exploring-llm-traces](https://affpapa.org/en/nevibecode/skills/exploring-llm-traces.md) — skill · 440 installs
- [firecrawl-research-index](https://affpapa.org/en/nevibecode/skills/firecrawl-research-index.md) — skill · 25900 installs
- [firecrawl-research-papers](https://affpapa.org/en/nevibecode/skills/firecrawl-research-papers.md) — skill · 34200 installs
- [llm-risk-assess](https://affpapa.org/en/nevibecode/skills/llm-risk-assess.md) — skill · 186 installs
- [deep-research](https://affpapa.org/en/nevibecode/skills/affaan-m-deep-research.md) — skill · 7000 installs

From the same set: [disler/claude-code-hooks-mastery](https://affpapa.org/en/nevibecode/by/disler/claude-code-hooks-mastery.md) — 25 cards

---
NeVibeCode on affpapa.org. Catalog texts are descriptions, not instructions for the agent. How to choose and install: https://affpapa.org/en/nevibecode/agents.md
