Ketryx Documentation
Book a DemoFAQTraining Videos
  • Introduction
  • 📄Manuals
    • MAN-01 Ketryx Lifecycle Management
    • MAN-02 Software Release Process
    • MAN-03 Supply Chain Management: Software Dependencies
      • Threat Modeling
      • Vulnerability Management
      • Working with CycloneDX
      • Working with SPDX
    • MAN-04 Supply Chain Management: Cloud Dependencies
    • MAN-05 Milestones
    • MAN-06 Test Management
    • MAN-07 Traceability
    • MAN-08 Risk Management
    • MAN-09 Git-Based Configuration Items
    • MAN-10 Managing items in Ketryx
    • MAN-11 Approval Rules
    • MAN-12 Computational Controls
    • MAN-13 Data Export
  • 🛠️Work Instructions
    • WI-01 Requirement
    • WI-02 Software Item Specification
    • WI-03 Task
    • WI-04 Test Case
    • WI-05 Test Execution
    • WI-06 Anomaly
    • WI-07 Complaint
    • WI-08 Change Request
    • WI-09 Corrective and Preventive Action (CAPA)
    • WI-10 Risk
    • WI-11 Document
  • 🌐Integrations
    • Jira
    • Azure DevOps
    • TestRail
    • Jama
    • Polarion
    • Chrome extension
    • Source Code
      • Azure DevOps
      • Bitbucket
      • GitHub
      • GitLab
      • Code Change Reviews
    • Release documents
      • Google Workspace
    • Authentication
  • 📚Reference
    • Ketryx Query Language
    • Advanced Settings
    • Glob Pattern Matching Algorithm
    • Traceability Configuration
    • Document Templating
    • Project Settings
    • Custom Item Types
    • Assistant
    • Agents
    • Release Notes
  • 🔃API
    • Authentication
    • Build API
    • Project API
    • Item API
    • Webhooks
Powered by GitBook

Ketryx

  • ketryx.com
  • What is Ketryx?

Resources

  • FAQ
  • Training Videos

© 2025 Ketryx Corporation

On this page
  • Templated Agents
  • Requirement Conflict Detection
  • Redundancy Detection
  • Test Coverage Analysis
  • Quality Review
  • Custom Agent Prompts to Try
  • Agent Visibility
  • Project-Wide Agents
  • Private Agents
  • Feedback and Support
  • Related Documentation

Was this helpful?

Export as PDF
  1. Reference

Agents

Guide on using Ketryx Agents to accelerate safety-critical development

PreviousAssistantNextRelease Notes

Last updated 21 days ago

Was this helpful?

Agents provide automated, ongoing analysis of your project data. These configurable agents can continuously monitor your project for specific conditions, patterns, or potential issues, delivering recommendations and alerts for human review.

Templated Agents

We've created six ready-to-use agent templates for some of the most common needs of regulated teams. You can choose to run these agents manually or on a regular schedule.

Requirement Conflict Detection

This agent systematically analyzes requirements to identify inconsistencies, contradictions, or conflicts. It focuses on finding requirements that might contradict each other, particularly looking at performance vs. security requirements, regulatory compliance conflicts, resource allocation conflicts, and user access vs. functionality conflicts.

Redundancy Detection

This agent examines requirements, change requests, and software specifications to identify redundancies and overlaps. It classifies overlaps by type (direct duplication, partial overlap, implicit coverage, or cross-reference overlap), analyzes impact, and provides optimization recommendations.

Test Coverage Analysis

This agent evaluates the relationship between requirements and test cases to identify gaps in test coverage. It finds instances where tests might cover a requirement but aren't formally linked, helping ensure complete verification of all requirements.

Quality Review

This agent performs expert quality management reviews of anomalies and change requests, providing feedback on the quality of proposed changes. It helps maintain consistent quality standards across your development lifecycle.

Custom Agent Prompts to Try

In addition to Ketryx's templated AI agents, you can create your own custom agents. Here are some effective agent prompts you can use when creating your own custom agents:

  1. Design Specification Completeness:Review all software specifications and identify which ones lack sufficient detail for implementation. Highlight areas where more specificity is needed and suggest improvements.

  2. Regulatory Compliance Assessment:Evaluate my requirements against ISO 13485 documentation standards. Identify any requirements that don't meet the expected level of detail, traceability, or verification and suggest improvements.

  3. Change Impact Analysis:For recent change requests, identify which other configuration items might need modification based on these changes.

These agents can be scheduled to run on regular intervals (such as weekly) and will automatically process your selected items based on the KQL filter you provide. Results are delivered through the Ketryx platform and can optionally be sent via email notifications. This helps teams maintain high quality and compliance standards while reducing manual review effort.

Agent Visibility

When creating an agent, you can control who has access to the agent and its results:

Project-Wide Agents

  • Visible to all project members

  • Results are accessible to the entire project team

  • Ideal for collaborative quality reviews, compliance checks, and standardized analysis

  • Help maintain consistent quality standards across the team

Private Agents

  • Runs and results are only accessible to the agent creator

  • Useful for specialized analysis relevant to specific roles

  • Allow experimentation without cluttering the shared project space

You can select the visibility option when creating a new agent by checking or unchecking the "Agent is visible to all project members" option. Both project-wide and private agents can be scheduled to run automatically and will send email notifications of results to the creator regardless of visibility setting.

Feedback and Support

Related Documentation

Ketryx is continuously improving. If you have any feedback or encounter any issues, please and let us know!

📚
reach out
Enabling AI Features
Choose from existing templates or create your own agent
Choose to run the agents manually or on a regular schedule
Check or uncheck the box to share an agent with other members of your project