New to Claude Skills? Learn how to install them →

mem0ai on GitHub

Memory Reviewer

Free

Enhance memory quality with actionable insights.

by mem0ai63k stars on mem0ai/mem0
1 views
Updated Aug 7, 2026
Get this skill

Free · Opens the source repo

What Memory Reviewer does

Memory Reviewer is a specialized tool designed to audit the quality of stored memories in your active project. It systematically identifies issues such as duplicates, contradictions, and low-confidence entries, providing a clear and concise summary of findings. This skill is particularly useful for developers and designers who rely on accurate and consistent memory storage to inform their decision-making processes and project management. By ensuring that memories are of high quality, users can avoid confusion and maintain a coherent understanding of their project context.

The skill operates by fetching all memories associated with the active project and grouping them by their metadata type. It scans each group for potential issues using defined detection methods, such as identifying near-duplicates based on noun overlap or flagging contradictions where opposing facts are recorded. Additionally, it highlights low-confidence entries and stale memories that have not been updated in over 180 days. The output is a compact summary detailing the number of duplicates, contradictions, low-confidence entries, untagged memories, and stale memories found.

When issues are detected, Memory Reviewer provides a list of specific memory IDs associated with each problem, allowing users to quickly locate and address the issues. The skill also suggests running the /mem0:dream command to consolidate duplicates and resolve contradictions, facilitating a more streamlined memory management process. This proactive approach to memory hygiene is essential for maintaining a robust and reliable memory system, especially after multiple memory writes or following health checks that indicate potential issues.

Memory Reviewer is an ideal tool for teams and individuals who frequently interact with memory storage and need to ensure its integrity. By using this skill, users can conduct periodic audits to maintain high-quality memories, ultimately leading to better project outcomes and more effective collaboration.

When to use it

Use this tool when you need to check the quality of stored memories, especially after multiple memory writes or health checks.

When not to use it

This skill is not suitable for modifying or deleting memories; it only reports findings without making changes.

What you can build with it

After Multiple Memory Writes

Run Memory Reviewer after a session with 5 or more memory writes to ensure quality and consistency.

Periodic Memory Audits

Use this skill for regular checks on memory quality to maintain clarity and avoid confusion in project context.

Resolving Conflicting Search Results

When search results seem conflicting, use Memory Reviewer to identify and address underlying memory issues.

How to install Memory Reviewer

View source

1. Install with the skills CLI

npx skills add mem0ai/mem0/memory-reviewer --agent claude-code

2. Or install it manually

Download the skill folder and drop it into ~/.claude/skills/ for all projects, or .claude/skills/ to scope it to one repo. Restart Claude Code so it picks up the new skill.

Anthropic's agentic coding CLI, and the reference implementation of Agent Skills. Drop a skill folder into ~/.claude/skills and Claude Code loads it automatically whenever a task matches the skill's description. Claude Code docs

Inside SKILL.md

Written by mem0ai

Memory Reviewer

Audits memory quality for the active project. Finds duplicates, contradictions, and low-confidence entries.

When to use

  • User asks "check my memories", "memory quality", "any duplicates?"
  • User runs /mem0:memory-reviewer directly
  • After a session with 5+ memory writes (suggest proactively)
  • After /mem0:health --deep identifies issues

Steps

  1. Fetch all memories for active project via get_memories with filters={"AND": [{"user_id": "<active_user_id>"}, {"app_id": "<active_project_id>"}]}, page_size=200. Paginate if needed — cap at 200 memories.

  2. Group by metadata.type. Common types: decision, convention, anti_pattern, task_learning, project_profile, user_preference, session_state.

  3. Scan each group for issues:

    IssueDetection method
    Near-duplicates>60% noun overlap within same type. Compare memory text after stripping stop words.
    ContradictionsOpposing facts about same topic (e.g., "use PostgreSQL" vs "use MySQL" for same component)
    Low-confidencemetadata.confidence < 0.3
    Missing typeNo metadata.type set
    Stalecreated_at older than 180 days with no updates
  4. Output compact summary:

memory-reviewer: project=<id> total=<N>
  duplicates:      <N> found
  contradictions:  <N> found
  low_confidence:  <N> found
  untagged:        <N> found
  stale:           <N> found
  1. If issues found, list them with memory IDs:
Issues:
  [duplicate] "<memory_a>" ≈ "<memory_b>" [mem0:<id_a>, mem0:<id_b>]
  [contradiction] "<memory_x>" vs "<memory_y>" [mem0:<id_x>, mem0:<id_y>]
  [low_conf] "<memory_z>" (confidence: 0.1) [mem0:<id_z>]
  1. Suggest action: "Run /mem0:dream to consolidate duplicates and resolve contradictions."

Constraints

  • Read-only — never modify or delete memories (that's /mem0:dream's job)
  • Max 200 memories per scan
  • Report findings, let user decide on action

Frequently asked questions about Memory Reviewer

Similar skills