Skillv1.0.0

ClawScan security

Surf Forecast Agent · ClawHub's context-aware review of the artifact, metadata, and declared behavior.

Scanner verdict

SuspiciousApr 28, 2026, 10:01 PM

Verdict: suspicious
Confidence: medium
Model: gpt-5-mini
Summary: The skill's purpose and code align (it queries Surfline and returns surf spot summaries), but the package metadata omits that the agent must execute the included Go program (requires the 'go' binary and network access), which is an important mismatch and operational risk you should understand before running it.
Guidance: This skill appears to do what it says (fetch Surfline data and produce a ranked top-3), but be aware it includes Go source that the agent is instructed to execute locally. Before running it: 1) ensure you have the Go toolchain installed and accept that 'go run' will compile and execute included source on your machine; 2) review the full Go source yourself (or have someone you trust do so) to confirm there are no hidden endpoints or unexpected behavior; 3) ensure outbound network access to services.surfline.com is acceptable in your environment; 4) if you do not want to execute code locally, ask the skill author for a remote, signed binary/release or a version that only uses HTTP calls from the agent (no local compile), and request that required binaries be declared in the manifest. The manifest's omission of the 'go' requirement and lack of integrity metadata are the primary coherence concerns.

Review Dimensions

Purpose & Capability: noteThe skill's name, description, SKILL.md, and Go code all consistently implement a Surfline-based surf-spot ranking tool. However, the registry metadata lists no required binaries while the SKILL.md explicitly instructs running `go run surfline_region_report.go` — so the manifest omits a genuine runtime dependency (the Go toolchain). Calling Surfline endpoints is expected for this purpose.
Instruction Scope: okSKILL.md confines the agent to: run the Go collector script, parse its JSON output, score spots locally, and return the top 3. It does not ask the agent to read arbitrary local files or unrelated environment variables, nor to send data to endpoints other than Surfline. The instructions emphasize that the Go script is data-collection only and scoring must happen in the agent, which limits scope.
Install Mechanism: concernThere is no install spec (instruction-only), which is low risk normally — but the included code must be executed via `go run`. The manifest failing to declare the 'go' binary (or any installation instructions) is an operational gap. Executing included source means arbitrary code runs on the host; while the visible code calls only Surfline endpoints, executing unreviewed code carries risk and the package provides no integrity metadata (hashes/signatures).
Credentials: okThe skill declares no required environment variables or credentials and the code uses only public Surfline endpoints. No unrelated secrets or config paths are requested. This is proportionate to the stated purpose.
Persistence & Privilege: okThe skill does not request always:true and uses default invocation settings. It does not attempt to modify other skills or system-wide settings in the SKILL.md or visible code.