Open-source AI memory platform that gives agents persistent long-term memory by ingesting data in any format and building a self-hosted knowledge graph, combining vector embeddings, graph reasoning, and ontology generation for meaning-based and relationship-based retrieval.
Open the source and read safety notes before installing.
Source-backed facts for citing this resource, derived directly from the registry — also available as plain text for AI assistants.
Decision playbook
Signals are present but mixed. Use the checklist below to confirm the source and operational safety for your environment.
0
78
—
No baseline selected
No major trust-signal divergence detected in the current selection.
Confirm ownership and provenance before trusting install instructions.
Source link availableRequired
Open the canonical repository and verify ownership.
Source provenance statusRequired
Marked as source-backed.
Metadata reviewed
Registry metadata indicates a reviewed listing.
Validate risk disclosures before installation or API wiring.
Safety notes presentRequired
Review the listed safety guidance before running commands.
Privacy notes presentRequired
Review data handling notes before connecting accounts or secrets.
Trust level risk gateRequired
Trust level does not block evaluation.
Check package metadata and artifact integrity signals.
Install payload available
Install or copy payload is available for review.
Package verification flag
No package verification flag provided.
Checksum metadata
No checksum provided for downloaded artifact.
Use compare context to validate trade-offs before adoption.
Compare tray has multiple entries
Add at least one more entry to compare trust differences.
Baseline comparison available
No baseline peer selected yet.
Diverging trust signals identified
No major trust-signal divergence found.
Adoption plan
Current risk score 16/100. Use staged verification before broader rollout.
Validate source and review signals before any execution.
Confirm source provenanceRequired
Source URL/provenance metadata is present.
Confirm metadata review state
Listing has review metadata.
Verify install payload
Install/config payload exists and can be inspected.
Confirm safety, privacy, and package integrity signals.
Review safety notesRequired
Safety notes are present.
Review privacy notesRequired
Privacy notes are present.
Verify package integrity metadata
No package verification/checksum metadata.
Adopt in controlled steps based on the selected plan.
Run in isolated sandbox firstRequired
Use a constrained sandbox and observe behavior across multiple tasks.
Roll out graduallyRequired
Roll out to a small cohort before wider usage.
Set monitoring and fallback
Define rollback path and monitor errors after adoption.
Evidence readiness
Required evidence gates are covered (5/6 signals complete).
Source repository/provenance is listed.
Required in this preset
Review metadata is present.
Required in this preset
Safety notes are present.
Required in this preset
Privacy notes are present.
Optional in this preset
Package integrity metadata is missing.
Optional in this preset
Install payload is available.
Required in this preset
Required evidence gates are covered for this preset.
Decision timeline
5/6 steps complete with no blocking gaps for this preset.
triage
Source/provenance metadata is available.
triage
Review metadata is available.
verify
Safety notes are available.
verify
Privacy notes are available.
verify
Package integrity metadata is missing.
rollout
Install payload is available.
No required blockers for this timeline preset.
## Editorial notes
Cognee is useful when Claude-adjacent teams want agents to build durable, connected memory from their own data rather than re-reading raw documents each time. It ingests data in many formats and continuously builds a self-hosted knowledge graph, so agents can recall facts by meaning and follow the relationships between them across sessions. Cognee also ships integrations, including a Claude Code plugin.
Its approach is distinct from conversation-memory tools: rather than extracting individual memories from chat turns, Cognee builds a knowledge graph over ingested data with vector embeddings, graph reasoning, and ontology generation, so retrieval is both semantic and relationship-aware.
## Key capabilities
- **Knowledge-graph memory** — ingests data and continuously builds a self-hosted knowledge graph for long-term, connected recall.
- **ECL pipelines** — an Extract, Cognify, Load flow that turns raw sources into graph and vector representations.
- **Graph plus vector retrieval** — combines vector embeddings and graph reasoning so results are searchable by meaning and by relationships.
- **Ontology generation** — cognitive-science-grounded ontology generation to structure and connect knowledge.
- **Multimodal ingestion** — unify data from various sources and formats into one knowledge layer.
- **Isolation and traceability** — tenant and user isolation, traceability, an OpenTelemetry collector, and audit traits for reliability.
- **Multiple clients** — Python (`cognee`), TypeScript, and Rust clients, plus a Claude Code plugin and other integrations.
- **Self-hosted** — runs locally with supported graph and vector databases.
## How teams use it
- **Company brain** — unify data from many sources into one knowledge layer that agents can query.
- **Persistent agents** — give agents long-term memory that connects facts and evolves as knowledge changes.
- **Graph-augmented retrieval** — retrieve by meaning and by relationships for more complex reasoning than plain vector search.
- **Cross-agent knowledge** — share a common knowledge graph across multiple agents.
- **Domain grounding** — ground agents in your domain data with ontology-structured context.
## Getting started
Cognee is open source and runs self-hosted. Install it with `pip install cognee` (TypeScript and Rust
clients are also available), configure a model provider and backing graph and vector stores, then
ingest your data so Cognee builds the knowledge graph through its Extract, Cognify, and Load steps.
Agents query the graph to recall facts and relationships with full context; a Claude Code plugin is
available for that workflow.
## Source notes
- The official repository describes Cognee as an open-source AI memory platform that gives agents persistent long-term memory by ingesting data in any format and building a self-hosted knowledge graph.
- Documented capabilities include combining vector embeddings, graph reasoning, and cognitive-science-grounded ontology generation so documents are searchable by meaning and connected by relationships that evolve over time.
- Cognee provides unified ingestion, graph and vector search, tenant and user isolation, traceability, an OpenTelemetry collector, and audit traits.
- Clients are available for Python (`cognee`), TypeScript (`@cognee/cognee-ts`), and Rust, and integrations include a Claude Code plugin.
- The GitHub repository is `topoteretes/cognee`, is Apache-2.0 licensed, is installed from PyPI as `cognee`, and targets Python 3.10+.
## Duplicate check
Checked current `content/tools/`, `content/mcp/`, agents, skills, hooks, rules, commands, guides, open pull requests, and repository-wide content for `Cognee`, `cognee`, `topoteretes`, `cognee.ai`, `github.com/topoteretes/cognee`, `AI memory platform`, and `knowledge graph memory`. Existing entries cover adjacent memory, RAG, and storage tools, but Cognee's knowledge-graph and ontology approach is distinct, and no dedicated Cognee tools entry, Cognee source URL duplicate, or open duplicate PR was found.
## Disclosure
Editorial listing. No paid placement or affiliate link is used.Cognee is useful when Claude-adjacent teams want agents to build durable, connected memory from their own data rather than re-reading raw documents each time. It ingests data in many formats and continuously builds a self-hosted knowledge graph, so agents can recall facts by meaning and follow the relationships between them across sessions. Cognee also ships integrations, including a Claude Code plugin.
Its approach is distinct from conversation-memory tools: rather than extracting individual memories from chat turns, Cognee builds a knowledge graph over ingested data with vector embeddings, graph reasoning, and ontology generation, so retrieval is both semantic and relationship-aware.
cognee), TypeScript, and Rust clients, plus a Claude Code plugin and other integrations.Cognee is open source and runs self-hosted. Install it with pip install cognee (TypeScript and Rust
clients are also available), configure a model provider and backing graph and vector stores, then
ingest your data so Cognee builds the knowledge graph through its Extract, Cognify, and Load steps.
Agents query the graph to recall facts and relationships with full context; a Claude Code plugin is
available for that workflow.
cognee), TypeScript (@cognee/cognee-ts), and Rust, and integrations include a Claude Code plugin.topoteretes/cognee, is Apache-2.0 licensed, is installed from PyPI as cognee, and targets Python 3.10+.Checked current content/tools/, content/mcp/, agents, skills, hooks, rules, commands, guides, open pull requests, and repository-wide content for Cognee, cognee, topoteretes, cognee.ai, github.com/topoteretes/cognee, AI memory platform, and knowledge graph memory. Existing entries cover adjacent memory, RAG, and storage tools, but Cognee's knowledge-graph and ontology approach is distinct, and no dedicated Cognee tools entry, Cognee source URL duplicate, or open duplicate PR was found.
Editorial listing. No paid placement or affiliate link is used.
Show that Cognee is listed on HeyClaude. Paste this Markdown into your README — it renders the badge and links back to this page.
[](https://heyclau.de/entry/tools/cognee)Cognee side by side with 3 alternatives on trust, install, platform support, and disclosed safety notes — all from reviewed registry metadata.
1 trust signal differ across this comparison (Submitter).
Next steps differ across entries — use the actions in the table below to copy install commands and source links per resource.
| Field | Open-source AI memory platform that gives agents persistent long-term memory by ingesting data in any format and building a self-hosted knowledge graph, combining vector embeddings, graph reasoning, and ontology generation for meaning-based and relationship-based retrieval. Open dossier | Open-source memory layer for AI agents and assistants that extracts, stores, and retrieves user, session, and agent memories so applications can personalize and remember across interactions, with Python and TypeScript SDKs and pluggable vector, graph, and key-value stores. Open dossier | Open-source Python multi-agent framework for building agent societies, role-playing agents, stateful ChatAgent workflows, RAG agents, synthetic data generation, MCP-enabled use cases, and research-scale agent experiments. Open dossier | Modular open-source Python framework for building AI agents and LLM workflows with structures, tools, memory, drivers, and RAG engines, from Griptape. Open dossier |
|---|---|---|---|---|
| Next stepsDiffers | ||||
| Trust | ||||
| Review status | ReviewedMaintainer reviewed | ReviewedMaintainer reviewed | ReviewedMaintainer reviewed | ReviewedMaintainer reviewed |
| Package trust | Package not verified | Package not verified | Package not verified | Package not verified |
| Source provenance | Source-backed | Source-backed | Source-backed | Source-backed |
| SubmitterDiffers | davion-knight | davion-knight | — | davion-knight |
| Install risk | Review first | Review first | Review first | Review first |
| Notes | Safety ✓ Privacy ✓ | Safety ✓ Privacy ✓ | Safety ✓ Privacy ✓ | Safety ✓ Privacy ✓ |
| Brand | ||||
| Category | tools | tools | tools | tools |
| Source | source-backed | source-backed | source-backed | source-backed |
| Author | topoteretes | mem0ai | CAMEL-AI | Griptape |
| Added | 2026-07-09 | 2026-07-09 | 2026-06-18 | 2026-07-09 |
| Platforms | CLI | CLI | CLI | CLI |
| Source repo | — | — | — | — |
| Safety notes | ✓Cognee ingests your data and builds a knowledge graph that agents read from, so treat graph content derived from external or user data as untrusted input and constrain what an agent may do based on it. Use tenant or user isolation so one tenant's knowledge is not retrieved for another, and review which sources are eligible for ingestion. The extraction and cognify steps call a model provider with ingested content; scope credentials to the minimum needed and keep them out of source control. Cognee runs self-hosted, so secure the graph and vector stores and their endpoints, and do not expose them on a public interface without protection. Keep production ingestion and permissions narrower than quickstart examples, and set retention rules for the accumulating graph. | ✓Stored memories influence future agent behavior, so treat memories written from external or user content as untrusted input and constrain what an agent may do based on retrieved memories. Scope memory by a per-user or per-agent identifier so one user's memories are not retrieved for another, and review which content is eligible to become a stored memory. The memory-extraction step calls a model provider with conversation content; use credentials scoped to the minimum needed and keep them out of source control. Self-hosting keeps memories in your own stores, while the managed platform processes them under its terms; choose the deployment that matches your data requirements. Keep production memory stores and permissions narrower than quickstart examples, and set retention rules for accumulated memories. | ✓CAMEL agents can coordinate multi-step tasks, call tools, use web/search integrations, connect to MCP examples, and run with provider credentials; review tool permissions before giving agents write access or account access. Large-scale agent societies and role-playing workflows can generate high volumes of model calls, tool calls, logs, synthetic data, and intermediate artifacts; set budgets, rate limits, and stop conditions before long runs. RAG, document, media, browser, communication, and data-tool extras may access local files, third-party APIs, vector stores, notebooks, or generated datasets; isolate experiments from production systems. CAMEL examples include MCP-oriented use cases, but MCP does not make connected tools safe by default. Scope server permissions, credentials, filesystem access, and approval gates separately. Do not treat generated code, generated datasets, citations, research summaries, or multi-agent decisions as verified until they have been reviewed against source data and policy requirements. | ✓Griptape structures (Agent, Pipeline, Workflow) can call tools that run code, execute shell or Python, query databases, scrape the web, call external APIs, and read or write files; review each tool's side effects before enabling it. Tool names, descriptions, schemas, rulesets, and retrieved documents become model-facing context, so treat them as untrusted input that can influence the agent. Off-prompt Task Memory keeps large or sensitive tool outputs out of the prompt by default, but data returned to the LLM, other tools, or downstream tasks still needs review. Human-in-the-loop approval, rate limits, timeouts, and rollback policies belong on any tool that performs account, billing, data, or infrastructure actions. Keep production permissions narrower than notebook or demo examples, and scope model-provider and tool credentials to least privilege. |
| Privacy notes | ✓Ingested documents and the resulting knowledge graph can contain personal, confidential, or proprietary data; apply retention, deletion, and access-control policies to the graph and vector stores. Extraction, ontology generation, and embedding send ingested content to the configured model and embedding providers, which process it under their own terms; local models keep that on your machine. Because Cognee is self-hosted, ingested data stays in the stores you configure, but any external providers used for processing follow their own data-handling policies. Provider keys, ingested data, and graph exports should be treated as sensitive and kept out of version control. | ✓Mem0 stores user preferences, facts, and history as memories, which can include personal or sensitive data extracted from conversations. Memory extraction and embedding send conversation content to the configured model and embedding providers, which process it under their own terms; local models keep that on your machine. Stored memories persist in the vector, graph, or key-value backend you configure, so apply retention, deletion, and access-control policies to that store. Provider keys, memory data, and any exports should be treated as sensitive and kept out of version control. | ✓Prompts, model responses, agent messages, tool arguments, tool outputs, retrieved documents, search results, logs, generated datasets, traces, and errors may include user or workspace data. Model providers, search providers, MCP servers, vector stores, web tools, document parsers, browser tools, and observability integrations may receive data from CAMEL workflows. Keep provider API keys, OAuth tokens, MCP server credentials, vector database URLs, generated logs, and synthetic datasets out of committed examples, screenshots, public issues, and shared notebooks. If `CAMEL_MODEL_LOG_ENABLED` or other logging/tracing integrations are enabled, review request/response logs and model configuration logs before sharing or retaining them. | ✓Griptape runs can send prompts, instructions, conversation memory, tool arguments, tool results, and retrieved context to configured model providers and embedding services. Tools and drivers can expose local files, database records, API responses, secrets, or proprietary business data to the model and workflow if they are made available to a structure. Conversation memory, task memory, vector stores, and any observability or run-logging destinations can retain prompts, outputs, embeddings, and metadata outside the application runtime. The managed Griptape Cloud processes data you send to it; review its data-handling terms before sending production or customer data. RAG and web/file loaders can pull third-party or workspace content into prompts, memory, and stored artifacts, so apply normal retention and access-control policies. |
| Prerequisites |
|
|
|
|
| Install | — | — | | — |
| Config | — | — | — | — |
| Citations | ||||
| Claim | Unclaimed | Unclaimed | Unclaimed | Unclaimed |
Open-source memory layer for AI agents with multi-level memory, extraction, and retrieval.
Python multi-agent framework for agent societies, ChatAgent workflows, RAG, tool use, MCP examples, data generation, and large-scale agent research.
Source-backed guides for putting this to work.
Master MCP server development from scratch.
Loading live community signals…
A short, calm digest of reviewed Claude resources. Unsubscribe any time.