graphiti-memory
Knowledge graph memory skill for haiku.skills using Graphiti and FalkorDB.
Enables agents to remember facts across conversations, recall relevant context, and forget outdated information.
Prerequisites
FalkorDB
Run FalkorDB via Docker:
docker run -p 6379:6379 -p 3000:3000 \
-v falkordb_data:/var/lib/falkordb/data \
falkordb/falkordb:latest
- Port
6379— Redis-compatible protocol (what the skill connects to) - Port
3000— Web UI for browsing the graph - The volume mount ensures data persists across container restarts
Ollama models
The skill uses Ollama for LLM and embeddings by default. Pull the required models:
ollama pull gpt-oss
ollama pull qwen3-embedding:4b
Configuration
All configuration is via environment variables:
FalkorDB
| Variable | Default | Description |
|---|---|---|
FALKORDB_URI |
falkor://localhost:6379 |
FalkorDB connection URI. Format: falkor://[user:password@]host:port[/database] |
LLM
| Variable | Default | Description |
|---|---|---|
OLLAMA_BASE_URL |
http://localhost:11434 |
Ollama server URL (shared with haiku.rag) |
GRAPHITI_LLM_MODEL |
gpt-oss |
Model for entity extraction and edge resolution |
GRAPHITI_SMALL_LLM_MODEL |
Same as GRAPHITI_LLM_MODEL |
Smaller model for lightweight LLM tasks |
Embeddings
| Variable | Default | Description |
|---|---|---|
GRAPHITI_EMBEDDING_MODEL |
qwen3-embedding:4b |
Embedding model name |
GRAPHITI_EMBEDDING_DIM |
2560 |
Embedding vector dimensions (must match the model) |
Multi-tenancy
| Variable | Default | Description |
|---|---|---|
GRAPHITI_GROUP_ID |
default |
Namespace for isolating memories between tenants/agents |
Tools
- remember — Store facts, observations, and context into the knowledge graph
- recall — Search the knowledge graph for relevant memories
- forget — Remove outdated or incorrect memories
Installation
uv add haiku-skills-graphiti-memory
Metadata
Release files for haiku-skills-graphiti-memory 0.8.0
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Source distribution (sdist)
| File | Size | Uploaded | Provenance | Actions |
|---|---|---|---|---|
| haiku_skills_graphiti_memory-0.8.0.tar.gz | 4.3 kB | Without provenance | Details for haiku_skills_graphiti_memory-0.8.0.tar.gz |
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Actions Reset |
|---|---|---|---|---|
| haiku_skills_graphiti_memory-0.8.0-py3-none-any.whl | Python 3 | none | any | Details for haiku_skills_graphiti_memory-0.8.0-py3-none-any.whl |
Total release size: 9.7 kB
Release files / haiku_skills_graphiti_memory-0.8.0.tar.gz
| Download URL | haiku_skills_graphiti_memory-0.8.0.tar.gz |
|---|---|
| Size | 4.3 kB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
2d773d145abcbe0a4119aa29b78bea611d1aeb84f9a9131626a37c4adda445e0
|
|
BLAKE2b-256 checksum How to use checksums |
71c48906a47a944d27914a32539b85888821ae1092f6cfb948c3c0f69a733825
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/6.2.0 CPython/3.13.12
|
Release files / haiku_skills_graphiti_memory-0.8.0-py3-none-any.whl
| Download URL | haiku_skills_graphiti_memory-0.8.0-py3-none-any.whl |
|---|---|
| Size | 5.4 kB |
| Tags | Python 3 |
|
SHA-256 checksum How to use checksums |
6e4f8d90367ab436d62bf704feea5cb90eadf26b85098b64eeab797362ae4de7
|
|
BLAKE2b-256 checksum How to use checksums |
50079343af95a553193eab9d32b5423e77c438b397b3e521676858e33ceac30b
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/6.2.0 CPython/3.13.12
|
Security for haiku-skills-graphiti-memory 0.8.0
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
No provenance found
None of the files in this release were uploaded with provenance metadata.
A note on package security
PyPI is a repository for distributing packages, not a service for auditing them. While third-party security researchers and our staff do their best to detect and remove malicious packages, PyPI cannot guarantee the quality, security or suitability of any given package.
To learn more about protecting yourself from malicious actors, visit our package security documentation.