Cognee

Standard adds OAuth data source integrations for Slack, Notion, Linear and Google Drive, and a GitHub integration re-indexes repositories on push, all reading content into memory, while native integrations wire Cognee into LangGraph, CrewAI, the Claude Agent SDK, the OpenAI Agents SDK, Google ADK, n8n, Claude Code and Codex. These read into Cognee or connect it to agents. No read-write or actionable routes that let an agent act in outside systems are documented.

SourceCognee, cognee.ai/pricing and docs.cognee.ai integrationsread 2026-09-21

Pipelines orchestrate Tasks into coordinated workflows with a typed runtime context, and developers can build custom tasks and pipelines; these sequence the steps that build and enrich memory. That is orchestration inside a single use case, memory processing, rather than control flow for the customer's agent workflows.

SourceCognee, docs.cognee.ai core-concepts/building-blocks/pipelines and guides/custom-tasks-pipelinesread 2026-09-21

Documents, code and application data, including Slack, Notion, Linear, Google Drive and GitHub sources, are ingested into a relational store of documents, chunks and provenance, a vector store of embeddings and a knowledge graph of entities and relationships, which recall queries by vector, graph or hybrid retrieval with provenance on answers and an optional dataset-level context index. That gives the customer a maintained index and graph over its knowledge that persists and stays queryable.

SourceCognee, docs.cognee.ai core-concepts/architecture and cognee.ai/pricingread 2026-09-21

A feedback system lets users rate answers to improve memory, and the permission system controls who can read and write datasets, but no approval step, consent checkpoint, escalation rule or pause for a person's sign-off before an agent acts is documented.

SourceCognee, docs.cognee.ai guides/feedback-system and full documentation exportread 2026-09-21

Multi-user mode enforces authentication and dataset-scoped permissions through users, tenants, roles and access control lists, with each tenant's data in a dedicated database, and the larger enterprise engagements add SSO. Cognee states it does not currently hold SOC 2, ISO 27001 or an equivalent third-party certification.

SourceCognee, docs.cognee.ai cognee-cloud/functionality/data-and-security and core-concepts/multi-user-mode/permissions-system/overviewread 2026-09-21

Agent session traces record what a decorated agent function did on each call, including calls that raised errors, and can be recalled later; Cognee emits OpenTelemetry spans and logs to an OTLP endpoint such as Langfuse, writes structured logs, and visualizes memory provenance. The traces documented cover Cognee's own operations and the decorated call, not a step by step view of the agent's prompts, tool calls, retrieved knowledge and outputs.

SourceCognee, docs.cognee.ai guides/agent-session-traces, setup-configuration/logging and integrations/opentelemetry-tracingread 2026-09-21

Sessions are short-term memory scoped to a user and session ID, holding recent queries, responses and the context used to answer them; session distillation turns guidance stated in a session into permanent lessons that later recalls respect, per-user preferences adjust retrieval for each user and agent, and forget can reset memory-only state or delete one dataset without touching others. Each memory layer has a stated scope and a stated lifetime.

SourceCognee, docs.cognee.ai core-concepts/sessions-and-caching, guides/session-distillation and core-concepts/main-operations/forgetread 2026-09-21

Cognee runs as managed Cognee Cloud, as open-source software self-hosted with Docker Compose, a Helm chart for Kubernetes, EC2 or Coolify, fully locally with Ollama and embedded stores, or as an Enterprise deployment in the customer's own cloud.

SourceCognee, docs.cognee.ai how-to-guides/cognee-sdk/deployment and cognee.ai/pricingread 2026-09-21

Examples and guides for building agents on Cognee's memory layer are published, but no catalog or set of ready-made agents or templates a customer adopts is documented.

SourceCognee, cognee.ai/pricing and docs.cognee.ai examples/overviewread 2026-09-21

Agents and applications call Cognee, a memory layer, through remember and recall, and no schedule, event or channel that starts an agent's work without a person or the calling application initiating it is documented.

SourceCognee, docs.cognee.ai (full documentation export)read 2026-09-21

The customer configures which LLM provider Cognee uses, from OpenAI, Azure OpenAI and Anthropic to local models through Ollama, and separately chooses the embedding provider, vector store and graph store. Model choice stays under the customer's control.

SourceCognee, docs.cognee.ai setup-configuration/llm-providers and embedding-providersread 2026-09-21

Cognee documents a REST API with an OpenAPI specification for datasets, ingestion, recall and activity, a Python API, a CLI and an MCP server for Cursor, Claude Code and other clients, and serve and push connect the SDK to Cognee Cloud or a remote instance. That is a stable API and SDK for Cognee's own platform.

SourceCognee, docs.cognee.ai api-reference/introduction, python-api and cognee-mcp/mcp-overviewread 2026-09-21

Cognee publishes memory benchmarks for its own engine, which measure the vendor's product, and its documentation describes no evaluation harness, scored test cases or quality gate a customer runs against its own agent.

SourceCognee, cognee.ai/pricing and docs.cognee.ai (full documentation export)read 2026-09-21

Web URL ingestion fetches page content with custom extraction rules to build memory, which is building the corpus rather than an agent operating a browser at run time, and no browser, desktop or computer control is documented.

SourceCognee, docs.cognee.ai guides/web-url-ingestionread 2026-09-21