RLM Tools
A persistent Python sandbox that keeps codebase exploration data server-side instead of dumping it into your context window. You get three tools: rlm_start to open a session on a directory, rlm_execute to run Python with built-in helpers like grep, read_file, and glob_files, and rlm_end to clean up. Variables persist across execute calls so you can grep for patterns, filter results, read matches, and build summaries incrementally. Only your print statements come back to the model. Benchmarks show 25-35% context reduction in typical workflows and up to 99% savings when exploring large codebases. Reach for this when your agent burns half its context budget reading files it never needed to see in full.
- GitHub stars
- 8
- tools
- 3
- transport
- STDIO
- last updated
- Feb 13, 2026
Tools
3 toolsPublic tool metadata for what this MCP can expose to an agent.
rlm_startOpen a session on a directory.
Open a session on a directory.
No parameter schema in public metadata yet.
rlm_executeRun Python in the sandbox.
Run Python in the sandbox.
No parameter schema in public metadata yet.
rlm_endClose session, free resources.
Close session, free resources.
No parameter schema in public metadata yet.
Sponsors
Agent CaffeineSponsoredClose your lid, come back to finished work
SnugSponsoredThe clipboard manager for the AI eraPromote your product hereReach 125k+ visitors monthly
WaxSponsoredBlock anything with supervision, without erasing your iPhonePromote your product hereReach 125k+ visitors monthly
knobsSponsoredSee your site on any phone, any day, in any languageReadme
RLM Tools
Your AI coding agent spends most of its token budget just reading your code — not reasoning about it. Every grep, file read, and glob result gets dumped into the conversation. On a large codebase, that's 25-35% of your context (and cost) burned on raw data the model never needed to see.
RLM Tools gives your agent a persistent sandbox to explore code in. Data stays server-side. Only the conclusions come back.
# Install in one line (Claude Code)
claude mcp add rlm-tools -- uvx rlm-tools
# Or Codex
codex mcp add rlm-tools -- uvx rlm-tools
That's it. Your agent automatically uses the sandbox for exploration. No config, no prompting changes.
What Changes
Without RLM Tools — agent greps for import UIKit, gets 500 matches dumped into context. Reads 10 files, burns all their content as tokens. Context window fills up. Agent forgets what it was doing.
With RLM Tools — agent runs the same exploration in a server-side Python sandbox. Data stays in sandbox memory. Only the print() output enters context:
Details
- Repository
- github.com/stefanoshea/rlm-tools
- Website
- github.com/stefanoshea/rlm-tools
- Package
rlm-tools- Transport
- STDIO
- Registry
- MCP registry, active