Skillv1.0.0

ClawScan security

marketing-psychology · ClawHub's context-aware review of the artifact, metadata, and declared behavior.

BenignMar 11, 2026, 3:24 PM

Verdict: benign
Confidence: high
Model: gpt-5-mini
Summary: This is an instruction-only marketing psychology reference; what it asks for and what it does are consistent with its documented purpose and it does not request credentials, installs, or out-of-scope access.
Guidance: This skill appears internally consistent and low-risk: it is a local reference and guidance document for applying psychology to marketing. Before enabling it, consider: (1) review the provided references/mental-models-catalog.md yourself (it is included) so you understand what the agent may cite; (2) do not place sensitive customer data or secrets into marketing-context.md (the skill will read that file if present); (3) the skill source/homepage is unknown—if provenance matters to you, verify the author or prefer a skill from a known publisher; and (4) although the skill does not require credentials or installs, avoid sending proprietary copy or PII into the agent unless you trust the execution environment.

Purpose & Capability: okThe name/description match the content: a catalog of mental models and runtime guidance for applying behavioral science to marketing. Nothing requested (no env vars, no binaries, no config paths) is unrelated to this purpose.
Instruction Scope: okThe SKILL.md confines runtime behavior to reading an optional marketing-context.md (for audience context) and the included catalog file under references/. It does not instruct the agent to read unrelated system files, call external endpoints, or access credentials.
Install Mechanism: okNo install spec or code files are present (instruction-only), so nothing is written to disk or fetched at install time.
Credentials: okNo environment variables, credentials, or external keys are required. The skill's minimal resource needs are proportional to its function.
Persistence & Privilege: okalways is false and the skill does not request persistent system privileges or attempt to modify other skills or system configuration.