Name: Evidence Research Memory Starter
Brand: Tetrees AI
Availability: InStock

Evidence Research Memory Starter preview 2

Evidence Research Memory Starter preview 3

Tetrees Agent guide

What this agent does

Research decisions become unreliable when evidence, inference, uncertainty, and owner preferences are mixed across chats and disconnected notes.

Designed for

Product researchers
Founders and strategy teams
Analysts who need auditable decision support

Inputs it works with

Research question and decision constraints
Source excerpts or factual notes
Owner-approved evidence and output preferences

Outputs you receive

Evidence map with confidence boundaries
Option comparison and decision memo
Reviewable growth proposals that remain inactive until accepted

How the workflow runs

1Separate supplied evidence from inference, assumptions, and unresolved uncertainty.
2Compare options against the owner constraints and produce an auditable evidence map.
3Draft a decision memo and propose any reusable preference for explicit owner review.

Example usage

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

Compare vendors with incomplete evidence

A product team must shortlist vendors while pricing, security, and migration evidence remains incomplete.

Example request

Compare these three vendors for a 50-person SaaS team. Separate sourced facts from assumptions and tell me what evidence to request next.

Expected outcome: A traceable comparison matrix, explicit uncertainty, follow-up evidence requests, and a decision recommendation tied to the stated constraints.

Turn research notes into a decision memo

A founder has accumulated mixed notes and wants a concise investment or product decision without losing the evidence trail.

Example request

Convert these notes into a one-page decision memo with evidence, inference, risks, confidence, and the next reversible action.

Expected outcome: A structured memo that preserves evidence boundaries, highlights material risks, and proposes a reversible next step.

A successful run should

✓ Every material conclusion is labeled as evidence, inference, or uncertainty
✓ Recommendations connect directly to owner constraints
✓ No growth proposal becomes active without owner approval

Boundaries and limitations

• It cannot verify sources that were not supplied through an allowed tool
• It provides decision support rather than legal, medical, or investment advice

Capabilities

researchevidence synthesisdecision support

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 v1.0.3

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.

Evidence Research Memory Starter