Back to skill
Skillv1.0.0
ClawScan security
Fromsoftware · ClawHub's context-aware review of the artifact, metadata, and declared behavior.
Scanner verdict
BenignApr 28, 2026, 10:05 PM
- Verdict
- benign
- Confidence
- high
- Model
- gpt-5-mini
- Summary
- This is an instruction-only, informational skill about FromSoftware that requests no credentials, installs no software, and contains no code — its requirements and instructions align with its stated purpose.
- Guidance
- This skill is informational and internally consistent: it neither installs code nor asks for credentials, so the technical risk is low. Consider that the skill's source/homepage is unspecified — if provenance or factual accuracy matters to you, ask the publisher for citations or prefer content from a known author. If you plan to use the skill in an automated agent, remember it can be invoked autonomously by default, but here that is low-risk because the skill has no extra access or install behavior.
Review Dimensions
- Purpose & Capability
- okThe name and description promise a developer profile; the skill is instruction-only and contains a lengthy, topical SKILL.md. It requests no binaries, env vars, config paths, or installs — everything requested is proportionate to an informational profile.
- Instruction Scope
- okSKILL.md provides research/summary guidance (including a read_when list) and contains only topical content about FromSoftware's history, design philosophy, timeline, and business. It does not instruct the agent to read local files, external credentials, or send data to third parties.
- Install Mechanism
- okNo install spec and no code files are present. Because this is instruction-only, nothing will be written to disk or executed beyond the agent using the SKILL.md text.
- Credentials
- okThe skill declares no environment variables or primary credentials and the instructions do not reference any secrets or unrelated services. Requested privileges are proportionate to an informational skill.
- Persistence & Privilege
- notealways is false (default) and model invocation is enabled by default — this is the platform norm. Because the skill has no credentials or install steps, autonomous invocation presents a low risk; still note that the skill's source is listed as 'unknown'.