Skip to content

Catalogagents

Cogni

situationalBlueshift — gaining momentum

MCP memory server that connects facts in an entity graph so agents reason over what they remember.

Visit getcogni.io →Add to kitOpen in Noemium

ledger entry · observed by @whysanesanders

Exposes memory tools over MCP with deterministic graph traversal rather than pure vector similarity, which helps with multi-hop questions that share no vocabulary with the answer. The free tier is genuinely useful, and it works with any model. Hosted-only for now; self-hosted deployments are on the roadmap.

Strengths

  • Deterministic entity-graph traversal reaches connected facts a vector search misses.
  • Works with any model over MCP; no model lock-in.
  • Free personal tier is genuinely useful.
  • No vector database or separate model to run.

Known limitations

  • Hosted service; memory text leaves your machine unless you wait for the self-hosted tier.
  • Effectiveness depends on the agent consistently calling recall before answering.
  • Pro and team pricing is not published transparently on the site.

Use it when

  • Agents that need multi-hop reasoning across disconnected facts.
  • Users who want persistent memory across sessions and clients.

Skip it when

  • You cannot use a hosted service; self-hosted deployment is still on the roadmap.
  • Your agent will not consistently call recall before answering.
  • Published Pro/Team pricing is required before you buy.

Buy-check

  • ·Price — freemium, Free personal tier; Pro for higher limits and graph inspection; Team/Enterprise custom as of 2026-08-19. Open the pricing page, not a screenshot.
  • ·Free tier — Free tier exists — burn it first.
  • ·Cancel fallout — What happens to your data/projects on cancel — check the ToS, not the FAQ.
  • ·Seat math — Per-seat pricing multiplies quietly. Count seats before the annual plan.
  • ·Who skips it — Also skipped when: You cannot use a hosted service; self-hosted deployment is still on the roadmap.; Your agent will not consistently call recall before answering..

Price trailchangelog →

No price move in this repo yet.

Facts

pricing
freemium1
price note
Free personal tier; Pro for higher limits and graph inspection; Team/Enterprise custom
free tier
yes
open source
no
api
yes
self-host
no
category
agents
evidence
source-verified

Sources

  1. 1getcogni.io[snapshot]

No source, no number.

Edit this page on GitHub

Similar joball alternatives →

toolverdictpriceossself-host
Cognisituationalfreemiumnono
Difyshipfreemiumnoyes
CrewAIshipfreemiumyesyes
LangGraphshipfreemiumyesyes

Facts from the catalog files — not scores you can buy. Quality and speed stay in the briefing, not in a fake 1–5 grid.

Dify
agentsship

Dify

Source-available LLM app platform with agent workflows and RAG.

APIFREE TIER

freemiumCloud Sandbox free; self-host Community free under restricted license; cloud Professional/Team paid yearly

Blueshift — gaining momentum
CrewAI
agentsship

CrewAI

Python framework for multi-agent crews — roles, tasks and processes.

OSSAPIFREE TIER

freemiumopen source (MIT); cloud Free Basic (50 exec/mo); Enterprise custom

Steady momentum
LangGraph
agentsship

LangGraph

Graph-based agent orchestration from LangChain — stateful and controllable.

OSSAPIFREE TIER

freemiumframework free open-source; LangSmith Plus $39/seat/mo

Blueshift — gaining momentum