Agents detail story

Claude Code should be shown as a shared execution lane with broad reach into files, tools, and model traffic, not just as the most common agent row.

The detail page needs to turn one agent version into an explanation of shared behavior, common power, and concentration risk.

This agent shapes the inventory story

Why does this agent deserve a dedicated story page?

Sessions

Highest volume in inventory

Unique users

Shared across several people

Sensitive file adjacency

NaN

Often present in auth and config work

Model spend

NaN

Concentrated enough to matter operationally

c6 detail reference

claude-code@1.2.0

Significance

Agent :

Claude Code

claude-code@1.2.0

Version :

v1.2.0

SDK Language :

js / node

SDK Runtime :

node

SDK Version :

0.38.0

Agent ID :

claude-code@1.2.0

Recent Sessions

c6 shows where the agent appears in recent live and ended work.

User	Endpoint	Model	Cost	Status
alice@company.com	alice-mbp.local	claude-sonnet-4-6	$0.43	active
carol@company.com	carol-mbp.local	claude-sonnet-4-6	$0.28	active
bob@company.com	bob-thinkpad.local	claude-opus-4-6	$4.10	ended

Users Running This Agent

The raw page also rolls the lane back up by person.

User	Sessions	Total cost	Last seen
alice@company.com	6	$2.14	today
carol@company.com	5	$1.82	today
eve@company.com	4	$1.40	today

Key findings

What this representative item changes in the story

3 findings

Finding	Body	Operator note
This agent is a platform lane	Its significance comes from repeated use across people, endpoints, and sensitive workflows.	That sharedness is what makes the lane strategically important.
Tool reach matters more than version count	The combination of shell, file, and network power is what raises consequence.	Execution shape matters more than raw adoption alone.
Baseline use can hide the interesting cases	Because the lane is normal, outlier behavior needs stronger framing to stand out.	Meaning-layer detail should surface those outliers directly.

Immediate actions

Point to action before deeper raw inspection

3 actions

Review the odd behavior pattern :

Separate routine developer use from the one session lane that changed the current story.

Map the shared tooling set :

Confirm which capabilities are normal for this lane and which should trigger review faster.

Link the lane to credential policy :

The dominant execution surface should have the strongest credential and endpoint expectations.

Agent execution map

The dominant agent family should be shown as a shared lane, not an isolated binary.

agent

Use the map to connect one agent version to the users, tools, files, and destinations that make it consequential.

Focus

claude-code@1.2.0

Dominant execution lane across recent developer work.

shared lane

Users

Who rides this lane?

alice@company.com

high-signal work

bob@company.com

normal baseline

Endpoints

Where does it run?

alice-mbp.local

sensitive repo context

bob-thinkpad.local

normal lane

Capabilities

What does it unlock?

bash

broad execution power

filesystem

sensitive file adjacency

External paths

Where can it send context?

api.anthropic.com

expected provider

secondary calls

review this path

users on lane

shared reach

high

boundary importance

elevated

next drill

agent detail

Evidence bundles

Bundle the story instead of reconstructing it from rows

4 bundles

Bundle	Body	Why it matters
Deployment bundle	Claude Code appears on several developer endpoints and in the most active recent sessions.	This is what makes it a shared execution lane.
Capability bundle	The lane regularly invokes bash, file operations, and model traffic together.	That combination is the operational story.
Sensitive-work bundle	Auth and configuration paths appear repeatedly adjacent to this agent family.	This is why the lane matters beyond install count.
Boundary bundle	The lane often sits next to outbound provider traffic and secondary calls.	That makes it central to boundary interpretation.