Reference

This page records stable user-facing semantics and points to authoritative interfaces for the installed version. Exact parameters belong to runtime schemas and source code.

Workspace

The default workspace is .reme/ under the launch directory. Override it at startup:

reme start workspace_dir=/path/to/workspace

The workspace uses the following top-level directories:

Directory	Meaning	User editing
`session/`	Source conversations used by memory workflows	Usually read-only
`mem_session/`	Generated agent-wrapper sessions and configuration	Do not treat as user memory
`resource/`	Imported source material	Add, replace, or delete as needed
`daily/`	Topic-named cards and date indexes produced from conversations and resources	Inspect and correct
`digest/`	Durable personal facts, procedures, and knowledge	Primary manual editing surface
`metadata/`	Derived indexes, graph, and catalogs	Do not edit manually

metadata/ can be rebuilt and must not be the only copy of memory. Backups should retain at least session/, resource/, daily/, and digest/. mem_session/ contains generated runtime state rather than durable user memory.

Markdown Memory

Memory files are ordinary Markdown with optional YAML frontmatter:

---
name: Build checks
description: Checks required before submitting code
---

# Build checks

Run pytest before submitting. This rule was recorded in [[daily/2026-07-13/build-checks.md]].

Paths are workspace-relative and file actions cannot escape the workspace.
ReMe can add a missing .md suffix where appropriate.
[[path/to/file.md]] is a workspace wikilink.
Wikilinks use literal workspace-relative paths; ReMe does not infer .md, search by filename, or resolve folder notes.
name and description are common summary fields. Treat other metadata as stable only when covered by the current schema.
Prefer ReMe move / delete actions for linked files so related links can be checked.

Configuration

Without config=, ReMe loads the built-in default configuration. A file passed through config= replaces that default and must contain a complete application configuration. Dot-notation arguments are then merged on top of whichever configuration was selected:

# Use the built-in default with a small override
reme start service.port=8181

# Use a complete custom configuration, then override its port
reme start config=/path/to/config.yaml service.port=8181

Common top-level groups include:

Setting	Meaning
`workspace_dir`	Workspace root
`timezone`	Timezone for daily notes and scheduled jobs
`service.backend`	`http` or `mcp`
`service.host` / `service.port`	Listen address
`service.transport`	MCP transport
`jobs`	Background, scheduled, and served actions
`components`	LLM, embedding, storage, and graph implementations

Embedding retrieval is disabled by default. Enable both components.as_embedding and components.embedding_store, then set components.file_store.default.embedding_store to default. The configured vector dimension must match the selected model. ${VAR} and ${VAR:-default} in string values expand from the environment. Never commit configuration or .env files containing secrets. Authoritative sources:

reme/config/default.yaml for default jobs, components, and parameters;
reme/schema/application_config.py for the top-level model;
reme/config/config_parser.py for merge, override, and environment rules.

CLI

The CLI has one uniform form:

reme <action> key=value nested.key=value

Common entrypoints:

reme help          # actions served by the current service
reme version       # installed version
reme health_check  # component status
reme status        # component memory estimates and process RSS
reme find_reme     # locate a running ReMe process
reme reindex       # clear and rebuild the file index
reme traverse      # inspect a wikilink neighborhood
reme graph_snapshot # return the category-rooted digest graph
reme chat          # stream a read-only workspace-agent conversation; requires an LLM

reme search supports start_date and end_date as inclusive YYYY-MM-DD filters. The live watcher indexes Markdown in daily/ and digest/; reme reindex additionally scans resource/ and JSONL. Use reme read for Markdown. For a non-Markdown text result, inspect it with reme stat and then reme load, because load returns the complete file. Auto Resource watches md, txt, json, jsonl, csv, yaml, and html files placed directly under resource/ or in resource/YYYY-MM-DD/. Auto Memory preserves source messages under session/dialog/ but omits recalled tool results and base64 data from the transcript. Generated daily filenames come from validated content names; provenance fields such as session_id, source_conversation, and source_resource remain the stable lookup identity. Actions other than start connect to the running service. Runtime reme help and reme/config/default.yaml are the authoritative action and parameter lists.

HTTP and MCP

The default HTTP service exposes served jobs as POST /{action}. While it is running, use:

/ for ReMe Studio when a web build is available;
/docs for interactive OpenAPI documentation;
/openapi.json for the machine-readable schema.

Start MCP with:

reme start service.backend=mcp service.transport=streamable-http

The default MCP URL is http://127.0.0.1:2333/mcp. MCP clients discover the current tool list on connection. Stream jobs such as chat are not exposed as MCP tools.

Compatibility and Changes

Python and dependency ranges: pyproject.toml
Request, response, and persistent structures: reme/schema/
Behavioral boundaries: tests/
Version changes: GitHub Releases

Get Started

Integration

Resources

Workspace

Markdown Memory

Configuration

CLI

HTTP and MCP

Compatibility and Changes

​Workspace

​Markdown Memory

​Configuration

​CLI

​HTTP and MCP

​Compatibility and Changes

Workspace

Markdown Memory

Configuration

CLI

HTTP and MCP

Compatibility and Changes