Skillv1.0.4

ClawScan security

Policy Risk Memo Architect · ClawHub's context-aware review of the artifact, metadata, and declared behavior.

BenignApr 22, 2026, 3:04 PM

Verdict: benign
Confidence: high
Model: gpt-5-mini
Summary: The skill is an instruction-only memo-writing assistant whose requested resources and runtime instructions are consistent with its stated purpose and do not ask for extra credentials, installs, or system access.
Guidance: This skill appears internally consistent and low-risk because it's instruction-only and requests no credentials or installs. Before installing, consider: (1) If you need source-backed memos, ensure the agent environment actually has browsing or API access to retrieve and cite sources—otherwise use reasoning-only or provide source documents. (2) Test with non-sensitive prompts first to confirm output style and evidence-handling. (3) Do not rely on this tool for legal or classified-intelligence decisions; it intentionally forbids those use cases. (4) If you want tighter control, disable autonomous invocation in agent settings so the skill runs only when you explicitly call it.

Purpose & Capability: okThe name and description (policy/risk memo authoring) match the SKILL.md instructions. The skill is instruction-only and asks for structured inputs and evidence modes that are coherent with memo production; it does not request unrelated credentials, binaries, or filesystem paths.
Instruction Scope: okThe SKILL.md strictly confines the agent to producing decision-focused memos, defining workflow, evidence categories, and required intake. It explicitly forbids activities like legal advice, intelligence-style certainty, inventing sources, or live verification claims. It does not instruct the agent to read system files, exfiltrate data, call arbitrary endpoints, or access environment variables.
Install Mechanism: okThere is no install specification and no code files; the skill is instruction-only, which minimizes on-disk risk. No packages, downloads, or third-party installers are declared.
Credentials: okThe skill declares no required environment variables, credentials, or config paths. All requested behaviors (evidence modes, clarifying questions, structured output) are achievable without additional secrets or system access.
Persistence & Privilege: okThe skill does not request always:true and has no special persistence needs. It does not instruct modifications to other skills or agent-wide settings. Autonomous invocation is allowed by platform default but the skill's capabilities do not raise additional privilege concerns.