Search and retrieve information from Graphiti Memory graph database. Covers search_memory_facts, search_memory_nodes, query construction, and group filtering. Use when user mentions Graphiti search, memory retrieval, finding past work, searching knowledge graph, or querying episodic memory.
/plugin marketplace add laurigates/claude-plugins/plugin install graphiti-plugin@lgates-claude-pluginsThis skill inherits all available tools. When active, it can use any tool Claude has access to.
Techniques for searching and retrieving information from Graphiti Memory. Provides proven search patterns for finding similar past work, error solutions, node summaries, and learning from historical data.
Automatically apply this skill when:
Facts Search: Find specific relationships between entities
Node Search: Get comprehensive entity summaries
Before starting, search for similar tasks:
# Search for similar operations
mcp__graphiti-memory__search_memory_facts(
query="FastAPI async database connection setup with PostgreSQL",
group_ids=["python_development", "agent_executions"],
max_facts=5
)
# Use results to:
# - Understand approach that worked before
# - Avoid past mistakes
# - Estimate time based on previous similar work
When to use: Before starting new work Value: Learn from past successes, avoid repeating mistakes
When encountering errors, search knowledge base:
# Search for similar errors
mcp__graphiti-memory__search_memory_facts(
query="PostgreSQL connection pool exhausted timeout",
group_ids=["error_resolutions"],
max_facts=3
)
# Use results to:
# - Apply known solutions
# - Avoid trying failed approaches
# - Understand root causes faster
When to use: When encountering errors Value: Faster resolution using known solutions
Get comprehensive view of entity relationships:
# Search for patterns across entity relationships
mcp__graphiti-memory__search_memory_nodes(
query="FastAPI performance optimization techniques",
group_ids=["python_development"],
max_nodes=5
)
# Use results to:
# - See all optimization patterns discovered
# - Understand relationships between techniques
# - Apply holistic approach
When to use: Need comprehensive understanding Value: See patterns and relationships
When you know a specific successful case:
# Search centered on a known good execution
mcp__graphiti-memory__search_memory_facts(
query="async testing patterns",
center_node_uuid="<uuid_of_successful_project>",
max_facts=10
)
# Use results to:
# - Find related successful patterns
# - See what worked in conjunction
# - Build on proven approaches
When to use: Building on known successes Value: Find related winning patterns
Search for specific types of knowledge:
# Search for user preferences
mcp__graphiti-memory__search_memory_nodes(
query="coding style preferences for Python",
entity="Preference",
max_nodes=5
)
# Search for established procedures
mcp__graphiti-memory__search_memory_nodes(
query="deployment workflow steps",
entity="Procedure",
max_nodes=5
)
# Permitted entity types: "Preference", "Procedure"
When to use: Looking for specific knowledge types Value: Targeted, relevant results
Good queries (specific, contextual):
Poor queries (vague, generic):
Use group IDs to narrow search scope:
# Search specific domain
search_memory_facts(
query="authentication implementation",
group_ids=["python_development"], # Only Python work
max_facts=5
)
# Search multiple related domains
search_memory_facts(
query="API authentication",
group_ids=["python_development", "nodejs_development"],
max_facts=5
)
# Search by activity type
search_memory_facts(
query="authentication errors",
group_ids=["error_resolutions", "security_audits"],
max_facts=5
)
# Search specific project
search_memory_facts(
query="deployment issues",
group_ids=["project_auth_api"],
max_facts=5
)
Facts contain relationships:
Entity A --[relationship]--> Entity B
Example: "FastAPI project" --[uses]--> "PostgreSQL async driver"
Look for:
Nodes contain entity summaries:
Entity: "FastAPI Authentication"
Summary: All relationships this entity has
Look for:
Search for similar tasks
search_memory_facts(
query="<your task description with tech stack>",
group_ids=["<relevant domain>"],
max_facts=5
)
Review results for:
Apply insights to current work
Search for similar errors
search_memory_facts(
query="<error type> <technology> <symptoms>",
group_ids=["error_resolutions"],
max_facts=3
)
Try known solutions first
Document if new solution needed
Search for similar successful work
search_memory_nodes(
query="<technology> <pattern> best practices",
max_nodes=5
)
Compare your approach to past successes
Note improvements for future episodes
Task: Build REST API with authentication
Search:
mcp__graphiti-memory__search_memory_facts(
query="REST API authentication implementation FastAPI JWT",
group_ids=["python_development", "agent_executions"],
max_facts=5
)
Results show:
- Past FastAPI + JWT implementation (SUCCESS)
- Lesson: "Use HTTP-only cookies for tokens"
- Pitfall: "Avoid storing tokens in localStorage"
- Time: Previous similar task took 90 minutes
Action: Apply proven approach, avoid known pitfalls
Error: "PostgreSQL connection pool exhausted"
Search:
mcp__graphiti-memory__search_memory_facts(
query="PostgreSQL connection pool exhausted timeout",
group_ids=["error_resolutions"],
max_facts=3
)
Results show:
- Root cause: Pool size too small for async workload
- Solution: Increase pool_size to 20, overflow to 10
- Verification: Load test after change
Action: Apply known solution immediately
Goal: Optimize FastAPI performance
Search:
mcp__graphiti-memory__search_memory_nodes(
query="FastAPI performance optimization techniques",
group_ids=["python_development"],
max_nodes=5
)
Results show relationships:
- FastAPI connects to async patterns
- Async patterns connect to connection pooling
- Connection pooling connects to performance gains
- Lessons about async context managers
Action: Apply holistic optimization approach
graphiti-episode-storage, graphiti-learning-workflowsgraphiti-memory (configured in settings.json)knowledge-graph-patterns (search sections)This skill should be used when the user asks to "create a hookify rule", "write a hook rule", "configure hookify", "add a hookify rule", or needs guidance on hookify rule syntax and patterns.
Create distinctive, production-grade frontend interfaces with high design quality. Use this skill when the user asks to build web components, pages, or applications. Generates creative, polished code that avoids generic AI aesthetics.