Overview

Shield360 ships first-class observability for AI coding agents. The shield360 CLI ingests events from each vendor’s hook system, normalizes them onto OpenTelemetry’s gen_ai.* and Shield360’s coding_agent.* semantic conventions, and ships them to your existing Shield360 collector. The Agents page’s Coding Agents tab detects them automatically - no controller, no SDK, no code change in your repos.

Note

You only need a Shield360 instance and the shield360 CLI on each developer’s machine. The CLI talks only to your Shield360 endpoint. There is no managed cloud for coding-agent telemetry.

What you get

Setup & Configure

Install the CLI, wire Claude Code, Cursor, or Codex, and stop tracking when you need to

Analytics

The seeded dashboard - sessions, cost, acceptance, top tools and repos, session outcomes

Sessions

Every coding session, with a full conversational drill-in per session

Users

Per-developer rollups of activity, cost, and code impact

Privacy & Governance

Content capture modes, redaction tiers, and attribution safeguards

Troubleshooting

Diagnose a missing row, silent hooks, or a stale database config

Scanner connectors

Scan the GitHub repo from vcs.repository.url.full with Trustabl and review findings on the session

Prerequisites

• A running Shield360 instance - see Installation if you don’t have one.
• A Shield360 API key (Settings → API Keys).
• The OTLP endpoint of your Shield360 collector (defaults to http://localhost:4318 for local installs).