01 / CONTINUITY
Knowledge that carries forward.
Retain decisions with their sources. Revise them as facts change. Retrieve the context in another session, with the history intact.
Retain · connect · revise
Sources and revisions remain inspectable.
ThinkingRoot is now Enkeon
Enkeon retains knowledge from documents, conversations and connected sources. People and AI tools can retrieve relevant context from the same project.
Mission
Make knowledge cumulative.
02 / SHARED CONTEXT
Context across your tools.
People and connected AI tools draw from shared project knowledge. Relevant context follows the work across supported MCP clients and the browser.
One project · connected tools
Access follows the project and branch scope.
03 / ENKEON-E
Experience becomes reusable.
Turn resolved conflicts into proposed procedures. Review them before reuse, so future work can draw on what the team has learned.
Feedback → review → reuse
Requires Enkeon-E access and supported clients.
From work to knowledge.
Connect your sources. Retain the decisions. Keep the evidence behind them.
Connected sources
Bring work into context
Connect supported apps and add documents to the project and branch where they belong.
Retained decisions
Keep the decision and why
Save useful knowledge from conversations, with its source. Revise it as the work changes.
01 / RETAINPROJECT / BRANCH
Files & documents
Connected apps
Saved conversations
ENKEON
Project knowledge
Sources · decisions · revision history
Knowledge in use.
Bring project knowledge to people and agents, across the tools they use.
People & AI tools
Recall where you work
Connected MCP clients and the browser extension retrieve relevant project knowledge when you need it.
Enkeon-E
Coordinate around context
Where enabled, participating agents share reported progress, questions and decisions within project scope.
02 / RETRIEVEPROJECT / BRANCH
AI tools · MCP
People · browser
Agents · Enkeon-E*
ENKEON
Project knowledge
Relevant context, retrieved when needed.
What you get.
Knowledge that carries forward. Context your tools can share. Experience you can build on.
01
Knowledge across tools
Files, conversations and decisions become project context for people and connected AI tools.
02
Continuity across sessions
Pick up with retained decisions and commitments after a new session, a tool change or compaction.
03
Answers with evidence
Inspect the sources and revision history behind retrieved knowledge. See what supports the answer.
04ENKEON-E*
Coordinated work
Share reported progress, questions and decisions so participating agents can account for each other’s work.
05ENKEON-E*
Learning from experience
Turn resolved conflicts into proposed procedures. Review them before reuse in future work.
06
Less context overhead
Retrieve relevant information for the task instead of repeatedly loading the entire project history.
* Enkeon-E · Coordination and learning require feature access. Availability depends on project configuration and supported clients.
What context changes.
Answer correctness, measured with and without retrieved context.
Without context
15%
3 / 20 correct
With Enkeon
85%
17 / 20 correct
Answer correctness14 additional correct answers
The cost of useful answers.
Token usage, model spend and execution time across the same 50 tasks.
Model spend / task
$0.37
49% lower · from $0.72
Cost / correct answer
$0.45
78% lower · from $2.01
Enkeon · leanBaseline = 100
Input tokens / task503,541 → 202,259
Evidence, in numbers.
Retrieval quality, memory integrity and latency. Each result, in its measured scope.
Recall latency
313ms
p50 server time
0400 ms
376 ms p95 · 60 queries.
Production image; server only.
Tokens processed
1B+
Reported total · 8 Oct 2026.
“It’s become Honeyb’s second brain.”
“Now it just knows which proposal we sent to which client, what we agreed and what’s still open.”
Matiss Katanenko · CEO, Honeyb
Email, meetings, and projects become a shared company memory in Enkeon. Both co-founders start with the same context in Claude or Codex.
Questions, answered honestly.
Every number on this page has a method page and a date. Same rule here.
What is Enkeon?
Enkeon retains knowledge from documents, conversations and connected sources. People and AI tools can retrieve relevant context from the same project, with source information and revision history.
How do I connect my AI tools?
Connect through MCP. The setup guide explains how to authorize the connection, select your tools and verify access to your project.
How is this different from keeping a file of notes?
Enkeon retrieves relevant knowledge for a question from retained project sources. It also retains source information and revision history. Files remain useful sources; the workflow comparison describes the particular static-context and retrieval configurations we tested.
What do the benchmark numbers mean?
Retrieval recall, generated-answer correctness and workflow cost measure different things. Our research pages describe each evaluation and its limits. Retrieved context supports an answer; it does not guarantee that the answer is correct.
How is access to project knowledge controlled?
Requests use project credentials and permissions. Projects, branches and user scopes determine which knowledge a request can access. Team access depends on the project's configuration.
Where is data processed?
Processing locations vary by service. The sub-processor list describes each provider, the data it receives and its processing location. The privacy policy explains data handling and international transfers.
From the blog
Research · Oct 23, 2026
Is memory optional for AI agents, or mandatory?
95% of enterprise AI pilots showed no profit impact, and bigger context windows don't fix it. The research, our own test (18 vs 41 of 50), and where the line sits.
Read the post →
Research · Jul 28, 2026
LongMemEval-S: measuring memory retrieval
An isolated retrieval evaluation: 99.0% Recall@15 on 499 scored questions, with 481 mean context tokens. Method, category results and limitations.
Read the post →