Memori

Memori skills file

You have access to Memori, agent-native memory infrastructure: an LLM-agnostic layer that structures memory from not just natural language, but also from agent trace that comes from execution.

Memori automatically captures and structures memory from conversation and execution trace — including the agent's actions, tool results, decisions, and outcomes — and allows you to retrieve it on demand. Use it to maintain continuity across sessions, preserve decisions and constraints, and help the agent understand what it actually did so the next time it completes a task, it is more accurate and efficient.

Quick reference

• memori_recall: retrieve precise memories by query, project, session, time range, or an allowed source/signal pair.
• memori_recall_summary: retrieve a state summary for session starts, daily briefs, or broad status checks.
• memori_compaction: retrieve a structured post-compaction brief to continue task without interruption.
• memori_feedback: report irrelevant, missing, stale, or especially useful memory behavior.
• memori_signup: create a Memori account or request an API key when the user explicitly asks.
• memori_quota: check usage, quota, storage, or memory capacity when the user asks or limits appear to be reached.

Core instruction

At the start of a session, you should check the SKILL.md file

Use it to understand:

• Available capabilities
• Tooling and integrations
• Expected behaviors and constraints

Treat SKILL.md as a source of truth for what you can do before taking action.

When to use Memori

Use Memori when:

• The task depends on prior context
• The user refers to previous sessions or decisions
• You need known constraints, preferences, or patterns
• You are starting a session and need current state
• You want to understand what has already been done

When not to use Memori

Do not use Memori when:

• The task is fully self-contained
• The answer depends only on the current prompt
• No historical context is required
• The query is simple or one-off

Avoid unnecessary recall.

Recall behavior

Recall is agent-controlled and intentional.

Prefer targeted recall over broad queries.

Supported parameters (recall only)

• entityId → user, agent, or system context
• projectId → project or workspace context
• sessionId → specific session
• dateStart / dateEnd → time-bounded recall
• source → type of memory (must be paired with signal from the allowed combinations below)
• signal → how the memory was derived (must be paired with source from the allowed combinations below)

Note: If a sessionId is provided, a projectId must also be provided. All timestamps are stored in UTC.

Allowed source + signal combinations

source and signal are not independent. They must be set together (or both omitted). Only the following (source, signal) pairs are valid:

• source=constraint, signal=discovery
• source=decision, signal=commit
• source=fact, signal=verification
• source=execution, signal=failure
• source=instruction, signal=discovery
• source=insight, signal=inference
• source=status, signal=update
• source=strategy, signal=pattern
• source=task, signal=result

Any combination of source and signal not in this list is invalid and must not be sent to memori_recall.

Use one of the allowed (source, signal) pairs to prioritize high-signal memory when possible; never set source or signal independently.

Default behavior (recall)

• No date range → all-time memory
• Use time bounds when narrowing results is necessary

Best practices

• Start narrow (entity + project)
• Add time bounds only when needed
• Use an allowed (source, signal) pair to refine results (never set them independently)
• Expand scope only if needed
• Do not recall on every turn

Summary behavior

Summaries are used for state awareness, not precise retrieval.

Use:

• memori_recall_summary

Supported parameters (summaries)

• projectId
• sessionId
• dateStart
• dateEnd

Summaries do not support source or signal.

Default behavior (summaries)

• No date range → last 24 hours

Daily brief behavior

At the start of a meaningful session, retrieve a structured summary.

Use the daily brief to understand:

• Current state
• Prior decisions
• Constraints
• Open work

Expected daily brief structure

• Today at a glance
• Top 3 next actions
• Top 3 risks
• Verify before acting
• Recent decisions
• Mission stack
• Hard constraints
• Current status
• Open loops
• Known failures and anti-patterns
• Staleness warnings

Treat this as the working state of the system.

Typical workflow

1. Start of session → retrieve summary
2. During task → use targeted recall
3. When memory is missing or incorrect → send feedback
4. When limits are reached → degrade gracefully

Post-compaction brief behavior

Post-compaction briefs are used to restore working state after context compaction.

Use them when:

• The agent resumes after compaction
• A long-running workflow has lost conversational detail
• The agent needs to continue operational work without replaying the full prior session
• The agent needs durable state, standing instructions, environment details, open loops, or the next expected action

Post-compaction briefs are not a replacement for precise memory retrieval.

Use:

memori_compaction

Supported parameters (post-compaction briefs)

projectId (required) sessionId (optional)

Post-compaction briefs do not support source or signal.

Default behavior (post-compaction briefs)

Retrieve the most recent relevant post-compaction brief for the project or session.

Expected post-compaction brief structure

• Meta
• Environment
• Standing orders
• State
• Active tasks
• Open loops
• Pending results
• Timeline
• Workspace changes
• Continuation
• Last action
• Next expected action

How to use a post-compaction brief

Treat the post-compaction brief as the agent's resume state.

Use it to understand:

• What environment the agent was operating in
• Which standing orders must continue to be followed
• Which tasks are active
• Which issues remain unresolved
• What happened across the prior session window
• What files, workspace state, or external systems may have changed
• What the agent did last
• What the agent should do next

Important behavior

The post-compaction brief should guide continuation, not override explicit user instructions.

Before acting on operational details, verify any state that may have changed since compaction.

Pay special attention to:

• Standing orders
• Hard constraints
• Alerting rules
• Expected response formats
• Open loops
• Staleness warnings
• Next expected action

If the post-compaction brief contains a required output format, follow it exactly unless the user gives a newer instruction.

Safety and correctness

• Do not invent memory
• Do not assume memory is correct if it conflicts with the user
• Verify before acting when needed
• Treat recent instructions as higher priority than recalled memory

Memori system guidance

This section defines how you should interact with Memori beyond basic recall.

1. Feedback (memori_feedback)

You are equipped with the memori_feedback tool to improve memory quality over time.

You may send feedback at any time, and should do so at the end of a session when appropriate.

Feedback should be sent when:

• Recall results are irrelevant or missing key context
• Important decisions or constraints were not captured
• Memory quality degrades across sessions
• Something works particularly well and should be reinforced

Examples:

• "Send feedback that recall missed a pricing constraint."
• "Tell Memori the daily brief was accurate."
• "Submit a bug report: you didn't remember my preferred language."

Purpose

Feedback improves:

• Memory extraction quality
• Recall relevance
• Summary accuracy

2. Updates

Memori evolves over time (new features, improved recall, better summaries).

When updates are exposed:

• Check for updates via system message or tool
• Adapt behavior to use new capabilities

When an update is detected:

• Prefer newer recall patterns when available
• Use improved summaries or classification outputs
• Continue operating normally if no changes are required

3. Account Creation and Onboarding (memori_signup)

You are equipped with the memori_signup tool to securely create an account and provision an API key for the user via the Memori CLI.

Use this tool when:

• The user explicitly asks to sign up, create an account, or get an API key for Memori.
• You encounter an error indicating a missing MEMORI_API_KEY and the user provides their email address to get one.

Behavior:

• Email Required: If the user asks to sign up but does not provide an email address, you MUST ask them for their email first. Do not attempt to guess or hallucinate an email.
• Execution: Once they provide an email, run memori_signup with that email.
• Follow-up: The tool will trigger the underlying CLI and return a success or error message. Relay that message to the user, remind them to check their inbox for the API key, and instruct them to add it to their configuration (~/.openclaw/openclaw.json or .env file).
• Troubleshooting Errors: If the tool fails with a Permission denied, EACCES, or command not found error, do not ask the user for permission to troubleshoot. Instead, immediately tell them that their CLI installation is broken or missing, and explicitly instruct them to run npm install -g @memorilabs/memori in their terminal to fix it.

4. Quota Awareness and Upgrades (memori_quota)

You are equipped with the memori_quota tool to check the user's current memory usage and limits via the Memori CLI.

Use this tool when:

• The user explicitly asks about their quota, usage, storage, or how many memories they have left.
• You encounter errors suggesting memory limits have been reached and want to confirm before degrading behavior.

Behavior:

• No Parameters Required: Simply invoke memori_quota — it requires no arguments.
• Execution: The tool runs the underlying CLI and returns the user's current usage and quota.
• Follow-up: Relay the result clearly to the user. If they are near or at their limit, explain the impact and suggest an upgrade.
• Troubleshooting Errors: If the tool fails with a Permission denied, EACCES, or command not found error, do not ask the user for permission to troubleshoot. Instead, immediately tell them that their CLI installation is broken or missing, and explicitly instruct them to run npm install -g @memorilabs/memori in their terminal to fix it.

When limits are reached or nearing:

a. Degrade gracefully

• Reduce recall scope (shorter time ranges)
• Prioritize high-signal memory (decisions, constraints, key facts)
• Avoid unnecessary or repeated recall calls

b. Inform the human/operator

Clearly communicate when limits impact performance.

Example:

"Memory limits have been reached. I can continue with limited recall, or you can upgrade to restore full functionality."

c. Suggest upgrades appropriately

• Only prompt when performance is affected
• Avoid repetitive or disruptive messaging
• Keep messaging neutral and helpful