rlm_start
Initialize a persistent sandbox session to analyze a codebase directory based on a query, returning compact summaries and metadata.
Instructions
Start an RLM exploration session. Returns session_id, metadata, limits, and available functions. Set include_guidance=true to include strategy/example coaching text.
Input Schema
TableJSON Schema
| Name | Required | Description | Default |
|---|---|---|---|
| path | Yes | Absolute path to the directory to explore | |
| query | Yes | What you want to find or analyze | |
| max_output_chars | No | Max characters per execute output | |
| max_llm_calls | No | Maximum llm_query/llm_query_batched calls for this session | |
| max_execute_calls | No | Maximum rlm_execute calls for this session | |
| execution_timeout_seconds | No | Per-rlm_execute timeout in seconds | |
| include_guidance | No | Include strategy/example guidance text in the response (larger payload) | |
| include_metadata | No | Scan directory and include file counts/types in response (set false for faster startup) |
Output Schema
TableJSON Schema
| Name | Required | Description | Default |
|---|---|---|---|
| result | Yes |