MCP Tools

memtomem exposes its LTM features as MCP tools. New users should keep the default core mode: it gives agents the common tools directly and routes everything else through mem_do, which keeps the agent’s visible tool list small.

Tool Modes

Set the mode with MEMTOMEM_TOOL_MODE in your MCP client config.

Mode	Tools exposed	Use when
`core` (default)	9 total, including `mem_do`	Best default for most agents
`standard`	38, including `mem_do`	You want common management tools directly visible
`full`	100 current tools	You are debugging, documenting, or using a client that handles large tool lists well

The following example is for a manual MCP-only connection. If a client plugin already supplies memtomem, check that client’s coexistence rule in Connect an AI Client before adding it.

Example:

{
  "mcpServers": {
    "memtomem": {
      "command": "memtomem-server",
      "env": { "MEMTOMEM_TOOL_MODE": "core" }
    }
  }
}

Core Mode

In core mode, the agent sees the tools needed for daily memory work:

Tool	Purpose
`mem_status`	Check server, storage, embedding, and index status
`mem_stats`	Return index statistics
`mem_add`	Store a new memory
`mem_search`	Hybrid search over indexed memory
`mem_recall`	Browse recent or date-filtered memory
`mem_index`	Index a file or directory
`mem_list`	List indexed memories / sources
`mem_read`	Read an indexed source file
`mem_do`	Route all non-core actions by name

Tell the agent what you want in natural language. It will usually pick the right core tool.

`mem_do`

mem_do provides access to every remaining action. It takes an action and optional params.

mem_do(action="help")
mem_do(action="help", params={"category": "context"})
mem_do(action="schedule_list")
mem_do(action="context_diff", params={"scope": "project_shared"})
mem_do(action="context_artifact_transfer", params={"asset_type": "skill", "name": "my-skill", "mode": "copy", "to_scope": "project_shared"})
mem_do(action="version")

Use mem_do(action="help") from your MCP client to see the action catalog for the installed version.

The v0.6.8 registry has 100 current tools and 94 mem_do actions. mem_context_migrate was removed and mem_candidate_evidence was added; the unchanged total does not imply unchanged membership. Pinned Context adds mem_pinned_list/get/set/delete and mem_context_compose; review-first formation adds mem_formation_scan and mem_candidate_propose/list/review/recover/evidence. The new full-mode surface also includes mem_context_pull for preview-first runtime imports and mem_quality_replay for deterministic Quality Lab replays.

mem_do(action="version") returns the server version, capability map, and a secret-free runtime_profile describing the effective embedding/rerank runtime and missing optional dependencies.

mem_candidate_propose creates a privacy-scanned pending review item rather than durable memory. It requires content, source, source_ref, and an idempotency_key; approved candidates alone pass through the normal write path. Reusing a key with identical content returns the original pending candidate; reusing it with different content is rejected.

mem_candidate_evidence compares a candidate with indexed memories in the caller’s scope, not other pending candidates or pinned context. A potential-conflict label is advisory, not a factual verdict or an automatic approval/rejection.

OpenCode

The published opencode-memtomem@0.3.11 plugin bundles Core 0.6.8. OpenCode uses the singular plugin key; there is no opencode plugin add command:

{"plugin": ["opencode-memtomem@0.3.11"]}

For MCP tools without the plugin’s slash commands and skills, configure a local server in opencode.json:

{
  "$schema": "https://opencode.ai/config.json",
  "mcp": {
    "memtomem": {
      "type": "local",
      "command": ["uvx", "--isolated", "--from", "memtomem[all]==0.6.8", "memtomem-server"],
      "enabled": true,
      "timeout": 60000,
      "environment": {"MEMTOMEM_TOOL_MODE": "core"}
    }
  }
}

An existing manual mcp.memtomem entry takes precedence over the plugin’s default, so only one server runs. Keep that exact key for a manual server, or remove the entry to use the plugin server. A different key such as mcp."memtomem-local" is not deduplicated and runs both. See the OpenCode connection guide for the full set of choices.

Common Actions

Category	Examples
Namespace	`ns_list`, `ns_set`, `ns_assign`, `ns_update`, `ns_rename`, `ns_delete`
Tags	`tag_list`, `tag_rename`, `tag_merge`, `tag_delete`
Sessions	`session_start`, `session_end`, `session_list`
Scratch	`scratch_set`, `scratch_get`, `scratch_promote`
Context Gateway	`context_detect`, `context_init`, `context_generate`, `context_diff`, `context_sync`, `context_pull`, `context_artifact_transfer` (move/copy skills, agents, commands across projects and tiers), `context_version`, `context_promote`
Maintenance	`dedup_scan`, `dedup_merge`, `decay_scan`, `decay_expire`, `cleanup_orphans`, `auto_tag`
Import / export	`export`, `import`, `import_obsidian`, `import_notion`, `ingest`
Quality / analytics	`quality_replay`, `activity`, `timeline`, `eval`, `reflect`

When to Change Modes

Stay in core when:

You are setting up memtomem for an agent.
You want lower prompt/tool-list overhead.
You are not sure which mode you need.

Use standard when an MCP client works better with direct tools than mem_do routing.

Use full when you are testing, writing docs, or manually exercising every tool. full intentionally exposes a large surface.

CLI Mirrors

Most common MCP operations have CLI equivalents:

MCP	CLI
`mem_status`	`mm status`
`mem_add`	`mm add`
`mem_search`	`mm search`
`mem_index`	`mm index`
`mem_do(action="context_sync")`	`mm context sync`
`mem_do(action="schedule_list")`	`mm schedule list`

If an agent is struggling to call the right tool, run the CLI command once yourself and then ask the agent to follow the same operation.

Full-mode tool names

These are the 100 current names in full mode, distinct from the nine tools exposed by default in core mode.

mem_activity
mem_add
mem_add_redaction_stats
mem_agent_register
mem_agent_search
mem_agent_share
mem_ask
mem_auto_tag
mem_batch_add
mem_candidate_evidence
mem_candidate_list
mem_candidate_propose
mem_candidate_recover
mem_candidate_review
mem_cleanup_orphans
mem_config
mem_conflict_check
mem_consolidate
mem_consolidate_apply
mem_context_artifact_migrate
mem_context_artifact_transfer
mem_context_compose
mem_context_detect
mem_context_diff
mem_context_generate
mem_context_init
mem_context_memory_migrate
mem_context_promote
mem_context_pull
mem_context_sync
mem_context_version
mem_decay_expire
mem_decay_scan
mem_dedup_merge
mem_dedup_scan
mem_delete
mem_do
mem_edit
mem_embedding_reset
mem_entity_scan
mem_entity_search
mem_eval
mem_expand
mem_export
mem_fetch
mem_formation_scan
mem_import
mem_import_notion
mem_import_obsidian
mem_importance_scan
mem_index
mem_link
mem_list
mem_ns_assign
mem_ns_delete
mem_ns_get
mem_ns_list
mem_ns_rename
mem_ns_set
mem_ns_update
mem_pinned_delete
mem_pinned_get
mem_pinned_list
mem_pinned_set
mem_policy_add
mem_policy_delete
mem_policy_list
mem_policy_run
mem_procedure_list
mem_procedure_save
mem_quality_replay
mem_read
mem_recall
mem_reflect
mem_reflect_save
mem_related
mem_reset
mem_schedule_delete
mem_schedule_list
mem_schedule_register
mem_schedule_run_now
mem_scratch_get
mem_scratch_promote
mem_scratch_set
mem_search
mem_search_feedback
mem_search_history
mem_search_suggest
mem_session_end
mem_session_list
mem_session_start
mem_stats
mem_status
mem_tag_delete
mem_tag_list
mem_tag_merge
mem_tag_rename
mem_timeline
mem_unlink
mem_watchdog