Back to skill
Skillv1.0.0
ClawScan security
Research Argument Builder · ClawHub's context-aware review of the artifact, metadata, and declared behavior.
Scanner verdict
BenignApr 25, 2026, 5:32 PM
- Verdict
- benign
- Confidence
- high
- Model
- gpt-5-mini
- Summary
- The skill is an instruction-only writing/argumentation helper whose requests and instructions are coherent with its stated purpose and do not ask for unrelated credentials or risky installs.
- Guidance
- This skill appears to be a straightforward, instruction-only research-argument assistant (status: draft). It's safe from an install/credential perspective because it asks for no binaries, downloads, or secrets. Before using, be aware that the skill will need access to your drafts or the research problem you provide (Read/Write tools) — avoid feeding sensitive PII or confidential data unless you're comfortable sharing it with the agent. If provenance matters, verify the repository/homepage and licensing on the linked GitHub page and prefer a non-draft/stable release for production use.
Review Dimensions
- Purpose & Capability
- okName, description, and declared inputs (a framed research problem / user's claim) align with the skill's actual instructions. The skill depends on the research-problem-framer and requires Read/Write tools to operate on user-provided text — behavior that is appropriate for an argument-building assistant.
- Instruction Scope
- okSKILL.md contains stepwise, domain-specific instructions about claims, reasons, evidence, warrants, and storyboarding. It does not instruct the agent to read system files, environment variables, or contact external endpoints unrelated to the task. Preconditions explicitly require a framed problem or invoking research-problem-framer, which keeps scope constrained.
- Install Mechanism
- okNo install spec and no code files — instruction-only — so nothing is written to disk or fetched at install time. This minimizes install-related risk.
- Credentials
- okThe skill declares no required environment variables, credentials, or config paths. The Read/Write tool access is proportional to the task of manipulating user-provided documents.
- Persistence & Privilege
- okThe skill is not always-enabled and does not request elevated persistence. It can be invoked by the agent (default), which is appropriate for a user-invocable writing helper and does not on its own indicate excess privilege.