Hook Claude Code
Runs before the agent uses a tool for Read, Grep, Glob, Bash, Read, Grep, Glob, Bash, Read, Grep, Glob and Bash tool calls, executing score.py, seal_run.py and rlm_repl.py (3 commands). From brainqub3/RLM.
Settings file Claude Code
Agent settings declaring 1 hook event (PreToolUse).
Skill Claude CodeCodex
Recursive Language Model (RLM) loop for processing a context that is too large to read into the conversation directly. Loads the context as a variable in a persistent Python REPL and answers the query by writing code that probes, chunks, and programmatically sub-queries a cheap LLM (llmquery) over slices of it, then…
Instructions file
Instructions for brainqub3/RLM, covering project instructions, headless codex sub-agent, rlm mode for long-context tasks, oolong eval — where run artifacts go and running the oolong rlm-vs-agent experiment.