RMS Memory MCP
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
No arguments | |||
Instructions
Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.
This server publishes no instructions, or was last inspected before Glama recorded them.
Capabilities
Features and capabilities supported by this server
Protocol revision2025-11-25
| Capability | Details |
|---|---|
| tools | {} |
| prompts | {} |
| resources | {} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| rms_searchA | Search RMS Memory. Returns {decision: inject|abstain, reason, injected_ids, results}; trust an abstain. corpus: vault (default, Markdown notes), code (derived semantic code), all (rank-fused). |
| rms_code_searchA | Search only the derived semantic code index. Results include language, file, symbol, kind, line range, and segment index. The code index is optional, so an unindexed project returns an empty result list. Pass |
| rms_readA | Read a markdown document from the RMS Memory vault by relative path (e.g. 'rules/api.md'). Whole note by default. For long notes save tokens: |
| rms_writeB | Save new architectural decisions, constraints, development rules, or project context to the RMS Memory vault. Use this tool PROACTIVELY at the end of a task if you learned a new user preference, solved a tricky bug, or made a new architectural decision. |
| rms_file_historyA | Query the derived code_path git file→commit history cache (no shell). Prefer this over |
| rms_graphA | Query and mutate the durable Markdown/code knowledge graph (MCP-first). Actions: status, ensure (reconcile vault links if empty or force=true), neighbors, path (BFS), snapshot, semantic (ephemeral embedding edges), create_edge / suppress_edge / edge_override (mutations require explicit |
| rms_doctorA | Run the vault health diagnostics (structure, IDs, links, LanceDB, wiki isolation, registry, freshness, credentials; plus pattern staleness when the vault has patterns/). Also reports a newer release (checked at most daily; RMS_MEMORY_NO_UPDATE_CHECK=1 disables) or a server binary replaced since start. Returns structured JSON. Set repair_frontmatter=true only with an explicit project (refuses sticky-bind repair). |
| rms_reindexA | Full rebuild of vault and/or code indexes. Destructive to derived tables — requires explicit |
| rms_syncC | Incremental vault index sync (same as CLI |
| rms_wiki_packA | Generate a wiki context pack from the vault and code index. Returns material for an agent to create human-readable wiki documentation from verified sources. |
| rms_projectsB | List registered RMS Memory project keys. This tool works even when the MCP client did not provide a workspace root. |
| rms_overviewA | Structured orientation summary for exactly one project: document counts by folder and status, recent notes, active checkpoints, index lag, and the code repository's git state (branch, HEAD, ahead/behind, changed file names; never diff contents). Call this at session start. Fail-closed: requires a bound workspace or an explicit |
| rms_pruneA | Find superseded vault notes older than a retention window and optionally archive them under artifacts/pruned/. Defaults to a dry run; set apply=true to archive candidates. Never deletes notes. |
| rms_hygieneA | Read-only memory hygiene report for one project: near-duplicate and identical notes (embedding similarity), active checkpoints untouched for 14+ days, and session notes older than 90 days, with next steps. Never edits the vault; merge duplicates with |
| rms_checkpoint_saveA | Create or update a session checkpoint (artifacts/checkpoints/.md, status=active). Save before context compaction or a long pause so work can be resumed. Updating preserves id/created_at and keeps omitted fields. The code repository branch and HEAD are recorded automatically. |
| rms_checkpoint_doneA | Close a checkpoint: marks it status=done (drops out of recall) and writes a durable session summary note under artifacts/sessions/. |
| rms_checkpoint_loadA | Load one checkpoint with its full body and bounded previews of linked notes. Use rms_checkpoint_query first to find names. |
| rms_checkpoint_queryA | List checkpoints for the current project, newest first, with full pending text. Filter with status=active|done|all (default all). |
| rms_remindA | Personal reminders that you pass on to the user. When the user says e.g. "remind me to call Viktor Petrovich tomorrow at 12", call action=add with |
| rms_system_instructionsA | Return the canonical RMS Memory usage protocol (search-first, persist, session continuity). Lets an agent self-bootstrap without injected rule files. |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
| session_start | Orient in the project memory: overview, active checkpoints, what to read first |
| save_progress | Save a resumable checkpoint and persist what was learned |
| record_decision | Record a technical decision without duplicating or contradicting existing notes |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
No resources | |
TDQS
Scored across 20 tools
Most tools target clearly distinct operations: search, read/write, checkpoint lifecycle, sync/reindex, diagnostics, graph, history, and reminders. A few pairs need careful reading to distinguish, notably rms_doctor vs rms_hygiene (diagnostics vs duplicate/cleanup reporting), rms_prune vs rms_hygiene, and rms_sync vs rms_reindex, but their descriptions explicitly clarify the boundaries.
All tools share a predictable rms_ snake_case prefix, which is consistent throughout. Within that, some names are verb-first (rms_search, rms_read, rms_write) and others are noun-first (rms_checkpoint_query, rms_file_history, rms_wiki_pack), but the style remains readable and mostly grouped by domain.
Twenty tools is on the heavy side for a memory server, though the product genuinely spans search, persistence, checkpoints, indexing, graph, diagnostics, and reminders. Tools like rms_system_instructions and rms_remind feel somewhat peripheral, making the surface borderline heavy rather than tightly scoped.
The surface covers search (vault/code/all), read/write, checkpoint create/query/load/close, project discovery, overview, incremental sync and full reindex, graph traversal/mutation, file history, diagnostics, hygiene, pruning, wiki packing, and reminders. Lifecycle coverage is broad and coherent, with only intentionally absent operations like hard deletion, which is explicitly replaced by pruning/archiving.