
Rlm Marketplace
- 1 repo stars
- Updated March 1, 2026
- ondrasek/spike-claude-code-rlm
rlm-marketplace is a Claude Code plugin that implements the Recursive Language Model pattern so developers can analyze documents larger than LLM context windows through recursive Python REPL exploration.
About
rlm-marketplace distributes a Claude Code plugin that adds Recursive Language Model analysis for documents far exceeding normal context windows. Instead of loading full text into the prompt, the LLM writes Python to inspect, search, chunk, and recursively sub-query an external CONTEXT variable in a REPL. Use /rlm or let Claude invoke it when analyzing large logs, codebases, or corpora. Requires uv on PATH and Python 3.11+; supports Anthropic, OpenAI-compatible, and local Ollama backends.
- Implements MIT CSAIL Recursive Language Model pattern (arXiv:2512.24601)
- Treats document context as external REPL variable, not prompt stuffing
- /rlm skill with multiple backends: Anthropic, OpenAI, Ollama, OpenRouter
- Bundled Python package executed via uv on first run
- Install via /plugin install rlm@rlm-marketplace
Rlm Marketplace by the numbers
- Data as of Jul 7, 2026 (Skillselion catalog sync)
/plugin marketplace add ondrasek/spike-claude-code-rlmAdd your badge
Show developers this marketplace is listed on Skillselion. Paste this into your README.
| repo stars | ★ 1 |
|---|---|
| Last updated | March 1, 2026 |
| Repository | ondrasek/spike-claude-code-rlm ↗ |
How can I analyze a document or corpus that is far too large to fit in a single LLM context window?
Analyze documents far beyond LLM context limits using Recursive Language Model exploration via /rlm and bundled Python REPL orchestration.
Who is it for?
Developers using Claude Code who need disciplined long-context analysis of logs, docs, or large text files via the RLM REPL pattern.
Skip if: Short documents that fit comfortably in one prompt or environments without uv and Python 3.11+ available.
What you get
RLM orchestration results synthesized from recursive inspect-search-chunk-subquery cycles over external context.
Plugins in this marketplace
1 plugin - install individually after you add the marketplace.
Recommended Marketplaces
FAQ
How is RLM different from RAG?
RLM keeps context as an external REPL variable and lets the LLM write Python to inspect and recursively sub-query it instead of retrieving chunks into the prompt.
What are the install commands?
Run /plugin marketplace add ondrasek/spike-claude-code-rlm then /plugin install rlm@Rlm Marketplace.
What runtime prerequisites are required?
uv on PATH and Python 3.11+; the plugin bundles the RLM package and resolves dependencies on first run.