graphiti-temporal
Skill12 starsChanged 29 days agoMITView on GitHub
Temporal context graph for agent memory — track entity relationships and state changes over time
In this repository22 other files it gives its agents.
- AGENTS.mdAGENTS.md
- docs/agents.mdAGENTS.md
- agent/CLAUDE.mdCLAUDE.md
- CLAUDE.mdCLAUDE.md
- router/CLAUDE.mdCLAUDE.md
- .github/copilot-instructions.mdCopilot instructions
- .cursor/rules/verification.mdcCursor rule
- agentic-agileSkill
- agentic-portfolioSkill
- client-onboardingSkill
- cooldown-resumeSkill
- drillSkill
- ecc-harness-patternsSkill
- fabric-patternsSkill
- managed-agents-dreamsSkill
- model-catalogue-updateSkill
- platform-setupSkill
- repowise-intelligenceSkill
- risky-module-reviewSkill
- seo-audit-reportSkill
- user-researchSkill
- video-contextSkill
The file
Files
Graphiti Temporal Context Skill
Inspired by: Graphiti — temporal context graphs for AI agents
Purpose: Integrate Graphiti's temporal knowledge graph patterns into local-llm-server's agent memory and context management.
What's Unique About Graphiti
Graphiti builds temporal context graphs (evolving knowledge graphs that track how facts change over time):
- Temporal awareness — knows what's true now vs what was true before
- Provenance tracking — maintains links to source data
- Hybrid retrieval — semantic + keyword + graph traversal
- Incremental updates — efficiently adds new information without full recomputation
Unlike traditional RAG (flat chunks), Graphiti gives agents rich, structured context that evolves with each interaction.
Integration Opportunities
1. Agent Memory as Temporal Graph
Track agent decisions and outcomes:
class AgentContextGraph:
def add_interaction(self, timestamp, agent_id, action):
"""Record agent action with temporal metadata"""
def query_at_time(self, entity, timestamp):
"""Query what was true at specific time"""2. Multi-Agent Coordination
Track which agents worked on which tasks with temporal awareness.
3. Knowledge Queries
Query across relationships with SQL:
SELECT entity, fact, timestamp FROM context_graph
WHERE entity LIKE 'test_%'
AND fact LIKE 'status:failed'
ORDER BY timestamp DESC;Database Schema
CREATE TABLE temporal_context (
id TEXT PRIMARY KEY,
entity TEXT NOT NULL,
fact TEXT NOT NULL,
timestamp DATETIME NOT NULL,
provenance TEXT,
agent_id TEXT,
created_at DATETIME DEFAULT CURRENT_TIMESTAMP
);
CREATE INDEX idx_entity_time ON temporal_context(entity, timestamp DESC);Files to Create
services/temporal_context.py— temporal graph implementationdb/temporal_store.py— SQLite storagetests/test_temporal_context.py— tests
References
- Graphiti: https://github.com/getzep/graphiti
- Quick-Note Issue: #263
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

