HmbownCommunity·8 Agent Skills Included

rlmagents

Recursive reasoning loops for large codebases and long documents

Runs recursive reasoning loops that analyze large codebases and long documents without overflowing the context window. Offloads big files into isolated sessions, then searches, chunks, and executes code against them instead of pasting everything into chat. Tracks evidence and citations across steps, compacts long reasoning trails, and ships with ready-made skills for SQL querying, research, and content writing.

50|4

↗

npx skills add Hmbown/rlmagents --all -g -y

Available:

Tells the agent to keep large payloads out of chat history by loading files into isolated contexts, searching and chunking them with dedicated tools, and finishing with evidence-backed answers.

All Skills in This Repository (8)

Pure Emerald Level Indicators

📦 In Repo

Hmbown

skill-creator

Guide users through creating and refining agent skills with structured directories.

arxiv-search

Searches arXiv preprint repository for papers matching user keywords and retrieves titles and summaries.

langgraph-docs

Retrieve and synthesize LangGraph Python documentation to answer user queries.

web-research

Plan structured web research, delegate subtopics to sub-agents, and synthesize findings into reports.

schema_exploration

List database tables and detail columns, data types, and relationships.

query-writing

Generate and execute SQL queries for data retrieval and analysis.

social-media

Generate platform-specific social media posts with images for LinkedIn and Twitter/X.

blog-post

Create structured, SEO-optimized blog posts with research and cover images.

Community

Intermediate

Frequently Asked Questions

FAQPage Schema

How to install rlmagents?▼

Run `npx skills add Hmbown/rlmagents --all -g -y` in your terminal to install all skills in this suite globally.

How to analyze large codebases without hitting context limits?▼

rlmagents loads big files into isolated sessions and lets the agent search, peek, and run code against them, so nothing huge is pasted into the chat window.

What is a Recursive Language Model loop?▼

It is a workflow where the agent writes and runs code in a session, spawns smaller sub-queries for hard parts, and finalizes an answer with tracked evidence instead of relying on one giant prompt.

Can rlmagents answer questions from a SQL database?▼

Yes. The included text-to-SQL example explores table schemas, writes safe read-only queries, and formats results from plain-English questions.

Does rlmagents work with Claude Code and other agent runtimes?▼

Yes. Its skills follow the standard SKILL.md format and run in Claude Code, Cursor, and other compatible agent environments.