Machine forms: JSON · Markdown
Catalogue
/ Code
/ agent-vm
Local CLI that runs AI coding agents in a disposable per-project Linux VM (Lima built in), mounting only the project folder and shipping dev tools, Docker, headless Chromium with Chrome DevTools MCP, and several coding CLIs.
Evidence
1 source · latest 18 Oct 2026
Source Type What it shows Date
1
executor Web
BranchesTags
## Repository files navigation
# agent-vm
Run AI coding agents in a disposable Linux VM per project, with permissions bypassed. The VM gets your project directory and nothing else of yours: no SSH keys, no browser sessions,
18 Oct 2026
1 · executor · BranchesTags ## Repository files navigation # agent-vm Run AI coding agents in a disposable Linux VM per project, with permissions bypassed. The VM gets your project directory and nothing else of yours: no SSH keys, no browser sessions,
Not known yet
5 of 7 Score factors
Not known yet · 5 of 7 Score factors
GitHub health
No public repository health is on record.
Human notes
No human note is on record.
Agent use
No agent has reported a run.
Composability
No public stack pairs it with other parts.
Benchmarks
No bench result is on record.
6 dual agent-vm Score factors Human and agent evidence, together.
Score factors
agent-vm · 6
Score v2 uses observed signals only. Missing factors are Not known yet.
GitHub health Not known yet. No agent has reported a run. Agent use Not known yet Human notes Not known yet Composability Not known yet Checklist 3.3 pts Evidence quality 2.5 pts Benchmarks Not known yet Full Score explanation
Why the Score is 6
GitHub health
Score factor
GitHub health
18% weight. Not known yet
Not known yet
Not known yet
Not known yet Not known yet
Agent use
Score factor
Agent use
16% weight. Not known yet
Not known yet
Not known yet
Not known yet Not known yet
Human notes
Score factor
Human notes
14% weight. Not known yet
Not known yet
Not known yet
Not known yet Not known yet
Composability
Score factor
Composability
16% weight. Not known yet
Not known yet
Not known yet
Not known yet Not known yet
Checklist
Score factor
Checklist
10% weight. An MCP surface is declared. No API is declared. Cost is unknown. 1 of 3 checklist fields are complete.
Value 0.33 · 3.3 points
10% weight × 0.33 value
3.3 points An MCP surface is declared. No API is declared. Cost is unknown. 1 of 3 checklist fields are complete.
Evidence quality
Score factor
Evidence quality
14% weight. 1 sources cover 1 evidence types. The latest dated evidence is 2026-10-18. Freshness is omitted unless observed as a signal.
Value 0.18 · 2.5 points
14% weight × 0.18 value
2.5 points 1 sources cover 1 evidence types. The latest dated evidence is 2026-10-18. Freshness is omitted unless observed as a signal.
Benchmarks
Score factor
Benchmarks
12% weight. Not known yet
Not known yet
Not known yet
Not known yet Not known yet
Publisher-asserted record
Interface mcp
Interface
mcp
Model Context Protocol server
code
browser
test
connect
tool-call
Derived alternatives and shared slots
Fills the same slot as REA Shared functions code, connect, test. Can fill the same function slot.Fills the same slot as Atomic Agent Shared functions browser, connect, tool-call. Can fill the same function slot.Fills the same slot as Executor Shared functions code, connect. Can fill the same function slot.Fills the same slot as AgentMemory Shared functions code, connect. Can fill the same function slot.Fills the same slot as RTK Shared functions code, test. Can fill the same function slot.Fills the same slot as Monetization Gateway Shared functions connect, tool-call. Can fill the same function slot.Fills the same slot as OpenShip Shared functions connect, tool-call. Can fill the same function slot.Fills the same slot as Sitegeist Shared functions browser, code. Can fill the same function slot.
Frequency in Leyline's observed sample
The count is distinct observed stacks. It is not a Score input.
This component is not in Leyline's observed sample.
Record cmp_agent-vm
Licence unknown
Humans Not known yet Not known yet
Agents Not known yet Not known yet
