Skillv0.1.1

ClawScan security

3dgs Experiment Planner · ClawHub's context-aware review of the artifact, metadata, and declared behavior.

BenignApr 30, 2026, 10:45 AM

Verdict: benign
Confidence: high
Model: gpt-5-mini
Summary: Instruction-only research assistant that recommends datasets, baselines, metrics, and ablation plans for 3D Gaussian Splatting; it requests no credentials, installs, or system access and its instructions are consistent with its stated purpose.
Guidance: This skill is instruction-only and internally consistent with its stated goal of designing experiments for 3D Gaussian Splatting. Before you use it: (1) avoid pasting proprietary datasets or private model checkpoints into prompts — the skill may echo or reorganize any text you give it; (2) verify any specific dataset or paper recommendations and licensing yourself (the skill may hallucinate or misattribute works); (3) treat outputs as suggestions to be validated by domain experts rather than authoritative facts; and (4) if you plan to let an agent invoke this skill autonomously, monitor its prompts and outputs during initial runs to ensure it does not propagate sensitive information.

Purpose & Capability: okThe name and description (3DGS experiment planning) match the SKILL.md content: guidance on datasets, baselines, metrics, ablations, and visualization planning. There are no unrelated required binaries, env vars, or config paths.
Instruction Scope: okSKILL.md contains researcher-style instructions and structured workflow (understand method, recommend datasets/baselines, propose metrics and ablations). It does not instruct the agent to read system files, access environment variables, call external endpoints, or exfiltrate data. (The provided SKILL.md is instruction-only and truncated in the listing but the visible sections are scoped to experiment design.)
Install Mechanism: okNo install spec and no code files — nothing will be written to disk or fetched during install. This is the lowest-risk install posture.
Credentials: okThe skill declares no required environment variables, credentials, or config paths. There is no disproportionate credential request relative to the skill's purpose.
Persistence & Privilege: okalways is false (default) and autonomous invocation is allowed (platform default). The skill does not request elevated persistence or system-wide configuration changes.