Agentic AI Atlasby a5c.ai
OverviewWikiGraphFor AgentsEdgesSearchWorkspace
/
GitHubDocsDiscord
iiRecord
Agentic AI Atlas · claim:codex-research-sandbox-modes
claim:codex-research-sandbox-modesa5c.ai
Search record views/
Record · tabs

Available views

II.Record viewspp. 1 - 1
overviewjsongraph
II.
Claim overview

claim:codex-research-sandbox-modes

Reference · live

claim:codex-research-sandbox-modes overview

Inspect the raw attributes, linked wiki pages, and inbound or outbound graph edges for claim:codex-research-sandbox-modes.

ClaimOutgoing · 5Incoming · 0

Attributes

subjectId
agent-runtime-impl:codex.runtime@1.x
subjectKind
AgentRuntimeImpl
attribute
subprocessSandboxStrategy
value
read-only, workspace-write, danger-full-access
statement
Codex CLI 0.120.0 exposes three `--sandbox` values: `read-only`, `workspace-write`, and `danger-full-access`. These modes are modeled as Codex-specific Sandbox nodes and launch recipes.
evidenceSourceIds
  • evidence:codex-research-cli-help-0-120-0
confidence
high
provenanceKind
observation
evidenceStrength
strong
status
accepted
claimedAt
2026-05-05T00:00:00Z
claimedBy
a5c-ai-catalog
claimKind
provenance
note
Supports Codex Sandbox nodes and --sandbox LaunchConfig recipes.

Outgoing edges

about_subject4
  • agent-runtime-impl:codex.runtime@1.x·AgentRuntimeImplCodex CLI Runtime (1.x)
  • sandbox:codex-read-only·SandboxCodex read-only sandbox
  • sandbox:codex-workspace-write·SandboxCodex workspace-write sandbox
  • sandbox:codex-danger-full-access·SandboxCodex danger-full-access sandbox
backed_by_evidence1
  • evidence:codex-research-cli-help-0-120-0·EvidenceSourceCodex CLI 0.120.0 help output

Incoming edges

None.

Related pages

No related wiki pages for this record.

Shortcuts

Open in graph
Browse node kind