Agent Skill

Rhesis ships one agent skill (rhesis) that covers the full workflow via the Rhesis MCP server — from requirements to test execution.

Workflow	When to use
Platform	Explore endpoints, design and run tests, analyze results
Requirements → test foundation	Paste a Product Requirements Document (PRD), product spec, or other requirements; create behaviors, metrics, tags, and test sets

One install. Invoke with /rhesis for a four-path menu when intent is unclear, or speak naturally and the skill routes automatically.

This is different from Rhesis’s inbound MCP connector (where the platform imports content from Notion, GitHub, or Jira). The agent skill lets an external AI interface call into Rhesis to drive testing workflows.

Four paths

On an ambiguous /rhesis (no endpoint, requirements doc, or run request yet), the skill offers:

Quick exploration — fast endpoint scan
Comprehensive exploration — full capability and boundary analysis
Requirements → test foundation — behaviors, metrics, tags, test sets from requirements
Run or analyze — execute test sets or compare past runs (returning users)

If you already paste requirements or name an endpoint, the menu is skipped.

→ AI agents: Agent reference and llms.txt → Terms: Glossary · Metric scope

Quick install

code.txt
npx skills add rhesis-ai/rhesis -g

Connect the MCP server in your agent — see Install in Cursor.

Examples


# Ambiguous — shows four-path menu
/rhesis

# Platform — explore and test an endpoint
"I want to test my travel chatbot. The endpoint is called 'travel-agent-v2'."

# Requirements — build a test foundation first
"Here's our agent PRD: [paste product requirements]"

# Returning user — run or analyze
"Compare my last two test runs for the safety test set"

Telemachus (native Architect)

The in-app Architect agent mirrors the same four paths. See Architect workflow.

Source

skills/rhesis/ — skill instructions and reference docs (references/prd/ for requirements design guides).