Skillv1.0.1

ClawScan security

x-cmd network skill · ClawHub's context-aware review of the artifact, metadata, and declared behavior.

Scanner verdict

SuspiciousMar 12, 2026, 10:40 AM

Verdict: suspicious
Confidence: medium
Model: gpt-5-mini
Summary: The skill claims to provide network admin/diagnostic functionality but its metadata and instructions are inconsistent: it requires external CLIs (x-cmd, nmap, curl) and elevated network privileges yet the registry lists no required binaries or install steps and the source provenance is minimal.
Guidance: This skill appears to be what it says (network diagnostics) but there are important gaps you should clear before installing or allowing it to run: 1) Confirm provenance — the registry metadata has no homepage and the source is 'unknown' even though SKILL.md lists a GitHub repo; verify that repository and its releases are trustworthy. 2) Ensure the required binaries exist and are from safe sources: x-cmd (the skill's CLI), nmap, and a curl-compatible tool are needed for many features — the skill should declare these in metadata. 3) Be cautious with privileges: some commands require administrator/root and raw socket access; run scans in a controlled environment and check your network policy/legal rules before performing scanning. 4) If you enable autonomous invocation for the agent, consider restricting or monitoring it: network scanning can be disruptive and has security/policy implications. If the publisher can clarify the missing required-binaries metadata and provide verifiable source artifacts, that would raise confidence; until then treat the package as suspicious and proceed only with verification and appropriate safeguards.
Findings: [no_code_files_instruction_only] expected: The static scanner found no code files or regex matches; this is expected because the package is instruction-only (SKILL.md). That means the runtime behavior depends entirely on the agent executing shell commands or calling local CLIs.

Review Dimensions

Purpose & Capability: concernThe SKILL.md clearly depends on an x-cmd CLI and on tools like nmap and curl (and sometimes root/administrator privileges) for its advertised functionality, but the registry metadata declares no required binaries and provides no install spec. That mismatch is unexpected: a network-tool skill would legitimately need those binaries and should declare them.
Instruction Scope: noteThe instructions direct running network-scanning and network-configuration commands (nmap scans, ARP inspection, DNS refreshes, routing table access, etc.) which are within the stated purpose. The SKILL.md does not appear to instruct reading unrelated files or environment variables, but it does require elevated permissions for certain operations (raw sockets, OS fingerprinting) — callers should be aware these operations can be intrusive on a network and may be restricted by policy.
Install Mechanism: okThis is an instruction-only skill with no install spec or code files, which reduces risk from arbitrary downloads. However, because it depends on external binaries (x-cmd, nmap, curl) the lack of declared required binaries in metadata is an inconsistency to resolve.
Credentials: okThe skill does not request environment variables, credentials, or config paths — which is proportionate for a CLI-based network diagnostic skill. There is no unexplained request for secrets.
Persistence & Privilege: okalways is false and the skill is user-invocable. Autonomous invocation is permitted by platform default but not combined with other high privileges here. The skill does require elevated OS/network privileges for some operations (e.g., raw packet scans), but it does not request persistent system changes or other skills' credentials.