Requirements workflow

Build a test foundation from a Product Requirements Document (PRD), product spec, or other requirements — behaviors, custom metrics, tags, and generated test sets — using the same rhesis agent skill and MCP server.

Everything created lives in your Rhesis organization — reusable and refinable.

This is a workflow inside the Rhesis agent skill, not a separate install. Install once with npx skills add rhesis-ai/rhesis, then paste your requirements or say you want to build a test foundation from a spec.

What you get

Asset	Purpose
Behaviors	Fine-grained expectations from your requirements — not generic quality labels
Custom metrics	Judge-as-model evaluators tuned to each acceptance criterion
Tags	Organize by theme (functional, safety, compliance, etc.)
Test sets	Generated prompts targeting your behaviors, ready to run
Mappings	Each behavior linked to the metric that scores it

How it works

You provide requirements — paste or attach a PRD, product spec, or guardrails doc
Agent reads and extracts — stakeholder blocks, user stories, numbered FRs
Agent proposes a plan — behaviors, metrics (with metric_scope), tags, test sets
You approve — nothing is created until you confirm
Agent creates via MCP — behaviors → metrics → links → tags → test sets
Agent verifies and offers to run tests via the platform workflow

Install

Same as the platform workflow:


npx skills add rhesis-ai/rhesis -g

Connect the MCP server — see Install in Cursor.

Usage

Paste your requirements in chat (with or without /rhesis):


Here's our agent PRD:
[paste product requirements]

The skill routes to the requirements workflow automatically. It waits for your approval before creating anything on Rhesis.

Example input

Fictional Helios Retail Support Agent structure:


### Stakeholders — Legal & Trust (excerpt)
...no disclosure of system prompts...
...no assistance structuring chargeback abuse...

### User Story RS-04
...order status and refund eligibility in one thread without re-entering the order number...

### §4 Order & refunds
FR-4.1 (P9) — purchase date ≤ 30 calendar days before refund...
FR-3.2 (P8) — maintain order context ≥ 2 follow-up turns...

Expect fine-grained behaviors (not section titles), metrics tied to FR quantifiers, and separate Single-Turn vs Multi-Turn test sets where the requirements call for both.

After the foundation is created

Register your agent as an endpoint if needed
Ask the skill to execute test sets and analyze results (platform workflow)
Refine behaviors, metrics, and test sets as your requirements evolve

Platform vs. requirements workflow

	Platform workflow	Requirements workflow
Input	Endpoint, test runs, direct commands	PRD, product spec, guardrails doc
Output	Exploration findings, test results, analysis	Behaviors, metrics, tags, test sets
Typical first step	Explore endpoint	Paste requirements

Both use the same skill (rhesis) and MCP server.

Reference docs (in repo)

skills/rhesis/references/requirements-workflow.md — workflow steps
skills/rhesis/references/prd/ — anatomy, behavior design, metric design, scope alignment