Conduct CLI is part of a platform focused on AI session governance, compliance, and automation for engineering teams. The platform addresses the challenges of monitoring and controlling AI agent actions, particularly in scenarios where agents can take real actions such as shipping code or running commands without oversight. It is designed for engineering leaders and teams who need visibility and control over AI usage, spend, and compliance across various AI tools.
Key features highlighted include governance of AI agent actions through policy enforcement, real-time spend limits, and a full audit trail that records who ran what, when, and why. The platform offers compliance frameworks aligned with standards such as the EU AI Act, NIST, ISO 42001, and OWASP. It provides automation packs and agent templates for tasks like PR review, security scans, CI failure triage, and more, which can run on multiple AI providers including Claude, GPT, and Gemini. The platform also supports integration with tools such as GitHub, Slack, Linear, Jira, and VS Code.
Conduct CLI is delivered as part of a broader solution with deployment options including SaaS, cloud, and on-premise. The platform is designed to work without requiring infrastructure changes and can be set up quickly. There is a free tier available, allowing teams to get started without initial cost. The platform aims to provide a single dashboard view of all AI tools, usage, and costs within an organization, helping teams maintain compliance and cost control while automating engineering workflows.
evidence_sufficient=false
In the AI security & guardrails space, conduct-cli takes a focused approach. It focuses on automating the installation, management, and testing of AI agents from the command line. It is built as an open-source project for AI developers and automation engineers. The project is open source (MIT). It runs on the web and the command line.
sseshachala builds and maintains conduct-cli, and it first shipped in 2026. Among its 4 catalogued features are agent management, project setup, and run tests.
Summary written by a language model from the project’s public pages.
What PulseGate has recorded for this listing
Closest matches by what these projects do