Jump To Category Page

Jump To Representative Detail

Agents Category View

Agent inventory should show which software families dominate execution, what they depend on, and where unfamiliar behavior begins to appear.

The category page should help an operator understand each agent as an execution lane shared across users, hosts, credentials, and tool surfaces rather than as a simple install count.

Use This Page When Asking

Which software lanes dominate execution behavior, and where is unfamiliar behavior entering them?

Representative Detail

claude-code@1.2.0

Open representative detail

Agent families

One dominant lane

Users reached

Mostly via one family

Unknown operator cases

Worth surfacing quickly

Version drift watch

new build pending

Compare against the stable baseline

Why This Page Exists

This category page should help the reviewer decide what matters in agents before the raw table takes over the reading experience.

What This Page Should Make Obvious

The reviewer should know what is routine, what is risky, and what deserves the next click.

Treat agent families as shared behavior surfaces

A dominant agent affects many sessions, keys, endpoints, and downstream surfaces at once.

Surface change signals early

New versions, error spikes, or new operators matter more than static install metadata.

Keep its runtime context visible

Models, credentials, and top tools explain what this agent really means to security.

Fastest Drill Paths

The next clicks should feel obvious and intentional.

Dominant coding agent lane

The main agent family appears across the widest set of users and assets.

Open representative detail

It is the fastest way to understand concentrated execution reach.

Raw version inventory

The list still needs a path to exact versions and supporting counts.

Open c6 raw list

Reviewers will still want the plain breakdown after the story framing.

Agent-Centered Topology

An agent map should connect the software to its users, hosts, dependencies, and change risk.

Useful per agent: who runs it, where it appears, which models and credentials it prefers, what capabilities it activates, and whether a new version or behavior spike stands out.

Agent

Claude Code v1.2.0

Dominant org-wide coding agent with the broadest footprint across users, endpoints, files, and MCP servers.

31 sessions · 12 users

Primary SDK

js / node

Common model

claude-sonnet-4-6

Common key

sk-ant-...A3F2

Version posture

stable · seen since Mar 15

Operators

Who this agent belongs to in practice.

3 nodes

alice@company.com6 sessions

carol@company.com5 sessions

unknown (10.0.1.44)1 session

Runtime Surfaces

Where the software has executed.

3 nodes

alice-mbp.localMacOS arm64

dave-mbp.localMacOS arm64

10.0.1.44unknown endpoint

Dependencies

What the agent tends to invoke.

3 nodes

claude-sonnet-4-6preferred model

filesystem MCPhigh-use server

read_file / bashtop capabilities

Watch For

Signals that change the meaning of usage.

3 nodes

new version arrivalupgrade event

error spiketool failures

unknown operatortriage path

c6 Raw Reference

c6 shape: compact language/runtime summary, map, then one row per agent family.

Languages

2 js/node · 1 python

Agents observed

Dominant lane

Claude Code

Last seen

today

Why Keep The Raw Ledger

c12 should still flow back into the raw inventory model. This page frames the question and likely answer; the raw table proves it.

Open c6 Raw List

Use the original category ledger for exact rows, counts, and timestamps.

Open c6 raw list

Raw Agent Inventory

Family, version, SDK language, sessions, and users.

Agent	Version	SDK language	Sessions	Users
Claude Code	v1.2.0	js / node	31	12
Cursor	v0.42.0	js / node	8	4
LangChain	v0.3.1	python	4	1