Name: Quant Research Council
Brand: Tetrees AI
Availability: InStock

Tetrees Agent guide

What this agent does

Investment research is often a single persuasive narrative with hidden assumptions, no as-of boundary, no competing analytical lenses, and no hard separation between a conviction signal and portfolio risk constraints.

Designed for

Independent market researchers
Quant-curious analysts
Portfolio teams preparing research memos
Builders of paper-trading and backtest workflows

Inputs it works with

Ticker or asset universe with an explicit as-of date
Fundamental, price, event, or owner-supplied evidence
Mandate, horizon, risk limits, and benchmark
Optional research notes or prior thesis attachment

Outputs you receive

Multi-lens signal ledger
Bull, base, and bear scenario matrix
Risk-clamped research allocation example
Evidence gaps and invalidation triggers
Research memo with a no-trade disclaimer

How the workflow runs

1Freeze the mandate, universe, horizon, benchmark, and as-of boundary before interpreting evidence.
2Evaluate the same evidence through distinct quality, value, growth, catalyst, and momentum lenses without pretending they agree.
3Convert each lens into a bounded signal with confidence, evidence, counterevidence, and missing-data flags.
4Synthesize the signals into scenarios and a research stance while preserving material dissent.
5Apply deterministic position and gross-exposure constraints after conviction; removed exposure stays in cash.
6Deliver an auditable memo and a validation plan for backtesting or paper research, never an autonomous live trade.

Example usage

Start with one of these requests, then replace the details with your own.

Research a concentrated equity thesis

An analyst has filings, valuation ratios, price history, and a six-month horizon but wants to avoid a one-story recommendation.

Example request

Assess NVDA as of 2026-06-30 for a six-month research horizon. Separate quality, valuation, growth, catalyst, and momentum signals, then show risk limits and thesis invalidators.

Expected outcome: A dated evidence ledger, independent lens scores, explicit disagreement, scenario probabilities, risk constraints, and a research-only conclusion rather than an execution instruction.

Challenge a long-short watchlist

A portfolio researcher wants to compare a small universe without letting the strongest narrative consume the whole risk budget.

Example request

Rank these six companies on a market-neutral research basis. Show signal confidence, missing evidence, pairwise risks, and what would falsify the top and bottom selections.

Expected outcome: A relative ranking with confidence bounds, missing-data penalties, deterministic exposure caps, and a reproducible paper-test plan.

A successful run should

✓ Every claim has an as-of boundary and evidence class
✓ Signals and risk constraints remain separate
✓ Dissent and missing data remain visible
✓ No live-trading action is initiated or implied
✓ Recommendations include invalidation triggers

Boundaries and limitations

• Research and education only; it is not investment advice
• It does not place orders or connect to a broker
• Market data must be supplied or obtained through an enabled search tool
• Backtests can overfit and do not predict future returns

Capabilities

quant researchfundamental analysisrisk controlscenario analysisdecision audit

Choose OpenAI, Claude or Z.AI in Agent Studio, then run with Tetrees Points or BYOK.

Execution tools

Hosted runs use only this signed set. Local MCP may add tools that you separately configure and approve.

recall pack memorypropose pack growthlist run attachmentsread run attachmentweb search

Agent AVCP

Agent AVCP report

Tested v1.0.0

The tested TAIP artifact is bound to this report and platform signature. Scores describe tested evidence, not a guarantee of every future model response.

Overall score

9.5

out of 10

Security9.5

Function10.0

Completion10.0

Mandatory gates

provenance✓ Pass

costCeilings✓ Pass

growthRollback✓ Pass

skillExecution✓ Pass

Reviews

No reviews yet.

Quant Research Council