Memory Memoria

Use Memoria as OpenClaw's durable memory slot. Triggers: "remember this", "save to memory", "what do you remember", "continue from last time", "forget this",...

Security Scan

The skill's requested actions and instructions line up with its stated purpose (durable memory via Memoria); it is an instruction-only plugin integration and does not request unexplained permissions or credentials, though setup examples reference embedding keys and a DB URL that you will need to provide to run it.

Purpose & Capability

The name/description (durable memory via Memoria) matches the runtime instructions: routing to memory_* tools, retrieval, store/correct/forget, snapshots, branches, and operational patterns. Nothing unrelated to memory management is requested.

Instruction Scope

SKILL.md stays focused on memory operations and includes explicit flows for retrieve/store/repair/rollback. The setup examples show CLI commands that install/verify the OpenClaw Memoria plugin. The instructions reference environment variables (DB URL, embedding provider, and embedding API key) in examples; they do not instruct reading unrelated files or exfiltrating data, but the examples imply the plugin will use external embedding services and a DB if you configure them.

Install Mechanism

No install spec or code files are provided by the skill itself — it is instruction-only. The install steps are user-invoked openclaw CLI commands (no automatic downloads or archive extraction by the skill). This is the lower-risk model for skills.

Credentials

The skill metadata declares no required environment variables, which is consistent with being instruction-only. However, the setup examples reference sensitive configuration (MEMORIA_DB_URL, MEMORIA_EMBEDDING_API_KEY / openai API key, embedding model and dims). Those are reasonable and proportionate for running a memory backend that uses embeddings, but they are sensitive credentials you would need to supply to run the plugin — the skill itself does not declare or require them up front.

Persistence & Privilege

always is false and the skill is user-invocable; it does not request permanent presence or attempt to modify other skills' configs. It describes creating snapshots/branches for memory state but does not request elevated agent privileges.

Assessment

This skill appears coherent for using Memoria as OpenClaw's long-term memory. Before enabling: (1) verify the plugin source code at the linked homepage and confirm you trust it; (2) be prepared to provide a database URL and an embeddings API key (the setup examples show OpenAI keys) — those are sensitive credentials and give the memory backend access to store and embed your data; (3) follow the skill's advice to snapshot before bulk deletes and to verify writes/repairs after changes; (4) consider privacy implications of storing secrets or personally identifying data in the external memory backend; and (5) if you only want ephemeral or file-based notes, prefer OpenClaw's built-in file memory rather than enabling this plugin.

Use this skill when OpenClaw should treat Memoria as the durable external memory system for the current user or project.

Routing

Pick the smallest reference that matches the task:

• Install or verify the OpenClaw plugin: references/setup.md
• Decide which memory tool to use: references/tool-surface.md
• Daily store, recall, correct, forget behavior: references/operations.md
• Session lifecycle, goals, recovery, branches, rollback: references/patterns.md

Core Rules

1. Prefer Memoria tools over MEMORY.md or memory/YYYY-MM-DD.md unless the user explicitly asks for file-based notes.
2. Do not auto-store every turn. Save durable facts, preferences, decisions, workflows, and meaningful progress.
3. On task resume or "what do you remember" prompts, retrieve relevant memory first.
4. Use the most specific tool available: memory_profile for stable preferences, memory_store for general durable memory, memory_correct or memory_forget for repairs.
5. Before bulk delete, purge, or large rewrites, create a snapshot first.
6. Use branches for risky or reversible memory experiments, then diff and merge or delete.
7. After important writes, repairs, rollback, or merges, verify with retrieval or list tools.
8. Do not claim only memory_search or memory_get exist when other memory_* tools are available.

Default Flow

1. At conversation start or task resume, use memory_retrieve or memory_search for relevant context.
2. During the conversation, store only the durable facts worth keeping.
3. If the user corrects or removes memory, repair it immediately with memory_correct, memory_forget, or memory_purge.
4. For risky memory maintenance, create a snapshot or branch before mutating state.
5. At the end of meaningful work, store the durable outcome and clean up obsolete working memory.

Important Notes

• OpenClaw's built-in file memory is separate from Memoria.
• The Memoria plugin defaults to explicit writes, not silent auto-capture.
• memory_get is a compatibility helper; when in doubt, prefer memory_retrieve, memory_search, or memory_list.
• Memoria's core strengths are semantic retrieval, durable cross-session memory, snapshots, rollback, branches, merge, and governance.

Quick Start

openclaw plugins install @matrixorigin/memory-memoria
openclaw plugins enable memory-memoria
openclaw memoria install
openclaw memoria verify