OpenClaw (first-party docs)Verified

BL-CLM-42.

OpenClaw — Human approval: Exec approvals (deny, allowlist, ask, auto, full) gate host command execution only and can be switched off; not a general approval layer.

Claim visualization

Actual claim data

BL-CLM-42 / Relationship

Claim evidence map

OpenClaw (first-party docs)

OpenClaw — Human approval: Exec approvals (deny, allowlist, ask, auto, full) gate host command execution only and can be switched off; not a general approval layer.

SECURITY.md says approvals are operator guardrails, not a multi-tenant authorization control. Covers host command execution, not every agent action.

Source: docs.openclaw.ai/tools/exec-approvals ↗
Where in the source: Where approvals are enforced; Ask modes; Fully unattended section
Source organization: OpenClaw (first-party docs)
Evidence type: Vendor documentation
Source date: Not shown on the source
Last verified: October 13, 2026
Approved wording: According to OpenClaw (first-party docs), exec approvals (deny, allowlist, ask, auto, full) gate host command execution only and can be switched off; not a general approval layer.
Scope and caveats: SECURITY.md says approvals are operator guardrails, not a multi-tenant authorization control. Covers host command execution, not every agent action.
Do not infer: Do not imply approval gates cover every agent action unless the source says so.

Cited onSelf-hosted AI agent platforms comparison ↗Papyrus vs openclaw ↗

← All claimsRecord BL-CLM-42