Back to skill
Skillv0.1.0
ClawScan security
Doctorbot Healthcheck Free · ClawHub's context-aware review of the artifact, metadata, and declared behavior.
Scanner verdict
SuspiciousFeb 12, 2026, 4:18 PM
- Verdict
- suspicious
- Confidence
- medium
- Model
- gpt-5-mini
- Summary
- The skill's stated purpose (a read-only health audit) mostly matches the commands it lists, but the instructions are vague about privileges, what data is collected or transmitted, and the SKILL.md expects local CLI tools without declaring them — these gaps merit caution before installing.
- Guidance
- This skill appears to be a read-only local audit and is instruction-only (no code installed). Before installing or running it, confirm: 1) whether the host has the referenced 'openclaw' CLI and other utilities the skill will run; 2) whether the 'openclaw security audit --deep' command requires sudo and what files it reads or transmits; 3) where the audit results are stored or whether they are sent off-host; and 4) that the agent will ask for your explicit consent before running any privileged actions. If you want safer usage, run the commands yourself interactively (or in a restricted environment) first, and avoid allowing automatic 'Pro' auto-fix actions without backups and explicit, per-action confirmation.
Review Dimensions
- Purpose & Capability
- noteThe name and description claim a local, read-only security audit and the SKILL.md lists commands consistent with that (openclaw security audit, update status, firewall/ports, OS/services). However the skill declares no required binaries or environment variables even though it expects an 'openclaw' CLI and may call system utilities (ufw, firewalld, port-listing tools). The omission of required binaries is a minor incoherence.
- Instruction Scope
- concernInstructions instruct the agent to run system-level commands (audit, firewall checks, port listings, OS/service queries). The SKILL.md repeatedly asserts the scan is 'read-only' and 'non-invasive' but does not describe: (a) whether any data leaves the host or is uploaded to an external service, (b) whether elevated privileges (sudo) are required for the 'deep' audit or firewall inspection, or (c) explicit confirmation prompts before running actions. The guidance 'The skill will guide you through the process' is vague and grants broad discretion to the agent.
- Install Mechanism
- okThis is an instruction-only skill with no install spec and no code files, so nothing is written to disk by the skill itself. That minimizes install-time risk.
- Credentials
- noteThe skill requests no environment variables or credentials, which is appropriate for a local audit. That said, it implicitly requires local CLI tools (openclaw, ufw/firewalld or equivalents) but those are not declared — the missing declaration should be clarified.
- Persistence & Privilege
- okalways is false and the skill is not requesting persistent elevated privileges. Autonomous agent invocation is allowed (platform default) but the skill does not request special persistence.