KLEIN.
Personal AI trading journal. v1 saves manual/paper records, computes local metrics and flags unchecked evidence. Generative AI is not connected.
Research / Define / Review
Caftra Labs.
Trading, with your own rules.
We’re building an international platform for AI trading on accounts users own. Start with KLEIN for your journal and KAVEN for your strategy rules, working locally today. Autonomous agents and broker execution are planned and remain disabled.
Functional Workspace v1 · Local tools · AI & broker execution disabled
Your first tools / Available locally
KLEIN + KAVEN.
Keep each trade connected to its reasoning, and turn your strategy into a structured blueprint with explicit risk limits. Save locally, review the evidence, and take your data with you.
Open the workspace to add your own records. The preview beside this text is a manually written illustration, not your saved data.
Open the local toolsOne lab, distinct products
MAVEN handles technical autonomy, AVEN fundamental and news autonomy, KAVEN user strategy automation, and KLEIN personal trade journaling. Customize Your AI is the planned agent builder.
Build in controlled steps
Local journal and strategy tools come first. Broker integration, an independent risk engine and live availability require separate testing, security and jurisdiction review.
How we develop our tools.
About Caftra LabsCAFTRA Labs / The product suite
Distinct roles.
One workspace.
A platform for traders using their own strategies and accounts. KLEIN and KAVEN are the first local tools; AI models, broker connections and every form of live execution remain inactive.
Control before autonomy.
KAVEN.
Automates user-defined strategies as the long-term goal. v1 structures and exports your rules and checks hypothetical risk; no bot runs.
MAVEN.
Autonomous technical trading AI, intended to learn from evaluated mistakes through controlled updates. Not implemented in this workspace.
AVEN.
Autonomous fundamental and high-impact news trading AI. No news feed, trading engine or autonomous execution is active.
Customize Your AI.
Agent builder to create a trading AI around the user’s workflow. No agent runtime or model training is available here.
A KLEIN review example.
Write your own recordA move beyond a prior trading range
Hypothetical observations, written to explain the process.
01 / Context
Location before direction.
A move means more when its location is explicit. Start by separating the market setting from the interpretation of a single observation.
Recorded observation
Price is testing the upper boundary of a prior trading range.
Still unresolved
The observation alone does not establish whether price will sustain outside that range.
Question to carry forward: what would distinguish a sustained move from a return into the range?
02 / Thesis
Make the expectation testable.
An upward move outside the range creates a continuation hypothesis. Write down what should follow and what would weaken that explanation.
Expected follow-through
Subsequent observations remain outside the range and sustain the upward move.
Weakening condition
Price returns into the prior range instead of sustaining outside it.
Question to carry forward: does the next evidence support the original expectation?
03 / Evidence
Compare the expectation with the evidence.
In this written example, subsequent observations remain outside the range. The original hypothesis has support, while uncertainty remains.
Observed in this example
Price remains beyond the prior upper boundary and sustains the initial upward move.
Interpretation to review
This is consistent with the continuation hypothesis. It does not establish a future outcome.
In this written example, price returns into the range. The evidence weakens the initial continuation hypothesis.
Observed in this example
Price moves back inside the prior range rather than sustaining outside it.
Interpretation to review
The expected follow-through did not occur. An opposite hypothesis would need its own evidence.
Question to carry forward: what changed in the evidence, and what changed in the thesis?
04 / Review
Keep the conclusion traceable.
Preserve the setting, the original expectation, and the observation that supported it. A human reviewer still evaluates the interpretation.
Research note
Continuation was supported by the observations in this illustrative case; uncertainty and weakening conditions remain explicit.
Human review
Check whether the recorded evidence supports that wording and whether relevant context is missing.
Retain the initial thesis alongside the evidence that weakened it. Revising an interpretation should leave a visible reasoning trail.
Research note
The continuation thesis weakened when price returned into the range. No opposite conclusion is established by that fact alone.
Human review
Check whether the revision follows from the evidence and whether a new hypothesis needs separate validation.
Question to retain: could someone else reconstruct this reasoning from the note?
Stage 1 of 4
This explorer contains manually written examples. It is a concept illustration, not a live assistant, AI response, market feed, or trading recommendation.
Planned journal workflow
The intended workflow is to add a research or trade note, organise its observations and assumptions, then review the supporting evidence and unresolved questions. The source note should remain available alongside the structured record.
Claude’s planned role
We plan to use the Claude API to organise user-written notes into structured records of context, assumptions, evidence and review questions. The application would provide the journal workflow around those outputs.
How we plan to evaluate it
Compare each generated note with its source record, check for invented or omitted facts, and flag unsupported conclusions. AI involvement and unresolved questions should stay visible.
What is available today
The product is at the design stage. You can explore manually written workflow examples on this website. The next intended milestone is a note-to-journal prototype, followed by evaluation against the source notes.
Research / Working questions
Evidence before
conviction.
These questions guide our first product’s development. Browse a topic and open a note to see the reasoning behind it.
Questions, not published findings.
Can an observation become a testable thesis?Connect the market setting to an expectation that can be checked.
A useful thesis connects context, a directional move, and an expected follow-through. It also states what would weaken or invalidate the interpretation.
We want the research process to preserve that sequence rather than reduce it to a single label. The workflow explorer illustrates how the same initial expectation can meet supporting or weakening evidence.
The question we’re exploring
Can a research note make the expectation and its failure condition clear enough for another person to review?
Explore the thesis stageCan higher-timeframe context stay intact?Keep the original setting visible when inspecting more detailed observations.
Lower-timeframe observations should support or challenge the original thesis without silently replacing it. A local move can look different when separated from the market structure around it.
Our research explores how to keep context and more detailed observations distinct and traceable. Both should remain available when the interpretation is reviewed.
The question we’re exploring
Can more detailed evidence be examined without losing the location and structure that gave it meaning?
Explore the context stageCan AI help review without hiding uncertainty?Ground explanations in supplied evidence and keep human review explicit.
Future KLEIN AI integration is intended to organise user-supplied research and journal entries. The output should preserve the original observations, assumptions, missing information, and review questions.
Evaluation will compare generated notes with their source records and check for invented facts, omissions, and unsupported conclusions. AI integration remains planned. The explorer is manually written; KLEIN’s local journal uses your saved records and rule-based calculations.
The question we’re exploring
Can AI organize an explanation while retaining its assumptions, missing evidence, and unresolved questions?
Explore the review stageThese are development questions and illustrative explanations. No performance results or completed studies are being presented.
About / Caftra Labs
Small beginnings.
Deliberate development.
A trading platform, built with deliberate controls.
Caftra Labs is an independent, unincorporated software project led by Melynkhael. We aim to build an international AI trading platform where users connect broker accounts they own. The public project began in October 2026.
The suite separates technical autonomy (MAVEN), fundamental/news autonomy (AVEN), user strategy automation (KAVEN), personal journaling (KLEIN), and an agent builder (Customize Your AI). Local journal and strategy tools are the first functional step.
Development note ·
Functional Workspace v1 adds local KLEIN records, KAVEN blueprints, risk calculations and reviewed backup import. AI models, broker connections, multi-user infrastructure and live execution are not active.
A few useful answers.
Will Caftra Labs build other products?
The suite is defined by role: MAVEN technical autonomy, AVEN fundamental/news autonomy, KAVEN strategy automation, KLEIN journaling, and Customize Your AI agent building. Only KLEIN and KAVEN local tools are functional in v1.
How will AI fit into the platform?
AI integration is planned for journal review and evaluated trading agents. v1 has no AI model or API connection: its summaries, metrics and prompts are rule-based. Trading execution would require an independent risk engine and separate security and legal review.
How does this differ from a general chat?
The workspace keeps strategy rules, journal records and source reasoning connected. You can edit and export structured records rather than lose them in a chat thread. AI assistance remains a future integration.
Are the products available?
KLEIN and KAVEN offer local journal and strategy-draft tools. Autonomous agents, broker connections, cloud sync and live trading are unavailable. The separate explorer contains illustrative examples.
Does the explorer generate AI analysis?
The explorer shows predefined, manually written examples. The functional workspace computes metrics and checks user-entered rules locally. Neither uses generative AI in v1.
Are there published trading results?
We are not presenting public performance results, backtests, or promises of returns. This website describes the project and its intended research approach.
Functional Workspace v1
Your work. In one place.
Record your trades, define your strategy, and review the evidence. This workspace runs locally in your browser.
The local workspace requires JavaScript. If it does not open, your saved data has not been changed.
Journal records0Your saved entries
Strategy drafts0Structured, editable rules
Closed records0Manual P&L entries
Win rate—Closed net-positive / all closed
KLEIN.
Your personal trading journal. Save context and execution, review net P&L by currency, and revisit missing evidence.
KAVEN.
Turn your strategy into a structured draft. Define the sequence and risk limits, then check a hypothetical paper plan.
Backup & restore.
JSON includes all saved journal records and strategies. CSV is for journal review. Exports do not include unsaved form changes.
Merge skips identical records and stops on conflicting IDs. Replacement requires confirmation. Local records are unencrypted, available to anyone with access to this browser profile.
Capability status.
- Local journal, calculations & draft rules: available
- Generative AI analysis: not connected
- Broker paper connection: not connected
- Live orders & broker credentials: disabled for every jurisdiction
- Cloud sync & multi-user accounts: not available
International access to a local workspace does not imply authorisation to offer live trading. Future availability depends on security, legal and broker review.
The rest of the suite.
Planned
MAVEN
Autonomous technical trading AI. Research, evaluation and controlled learning before any execution service.
Planned
AVEN
Autonomous fundamental and high-impact news trading AI. No news feed or decision engine is connected here.
Planned
Customize Your AI
Agent builder for a user's trading workflow. No agent creation or runtime in this release.
Local tools use calculations and rule-based prompts, not an AI model. No returns, strategy validity or execution quality are inferred from journal outcomes.
KLEIN / Personal AI trading journal
Every trade. With its context.
Capture what you knew, what you expected, and what happened. v1 provides local records and rule-based review; AI is not connected.
Enable JavaScript to use the local journal.
New record.
A manual record does not connect to a broker. Prices and risk are entered by you.
Your journal.
Start with one record.
Your journal is empty. Add a trade or import your CAFTRA backup.
| Trade / UTC | Status | Net P&L | Net R | Actions |
|---|
Evidence to revisit.
Rule-based counts from your selected records. These flags cannot verify the chart or diagnose why a trade won or lost.
Statistics use only closed records and keep account currencies separate. Entries are self-reported; small samples do not establish an edge.
KAVEN / Strategy builder
Your rules. Made explicit.
Define the sequence and risk constraints for your own strategy. v1 saves and exports a blueprint; automation remains disabled.
Enable JavaScript to use the local strategy builder.
New strategy.
Descriptions require human interpretation. Saving a strategy does not validate it or run a backtest.
Saved blueprints.
Use a blueprint in KLEIN to link a new record. Removing a linked strategy requires unlinking its records first.
Check a paper plan.
A hypothetical risk calculation from your saved rules and inputs. No market feed, signal, broker sizing or order submission.
The preview never reserves risk, updates a balance or records an order. Recalculate when inputs change.