Back to skill
Skillv1.0.0
ClawScan security
social-content · ClawHub's context-aware review of the artifact, metadata, and declared behavior.
Scanner verdict
SuspiciousMar 11, 2026, 3:51 PM
- Verdict
- suspicious
- Confidence
- medium
- Model
- gpt-5-mini
- Summary
- The skill's content-creation purpose matches its instructions, but the runtime instructions reference a local workspace file and give explicit guidance for large-scale scraping using third-party tools (Apify, PhantomBuster) without declaring those access needs—this mismatch merits caution.
- Guidance
- This skill appears to be a legitimate social-media content guide, but it asks the agent to read a local file (.claude/product-marketing-context.md) and includes instructions for large-scale scraping using third-party tools. Before installing or enabling it: (1) Confirm you are comfortable with the agent reading .claude/product-marketing-context.md or remove/sanitize that file if it contains sensitive info. (2) Be aware that following the reverse-engineering guidance could lead to mass scraping or require third-party credentials; do not provide credentials unless you trust the source and understand legal/ToS implications. (3) If you need tighter control, restrict the agent's file access or run the skill in an environment where it cannot reach scraping tools or the public internet. (4) If unsure about the publisher (source/homepage unknown), consider asking for provenance or using a vetted alternative.
Review Dimensions
- Purpose & Capability
- okName and description (social content creation, scheduling, optimization) align with the SKILL.md and supporting reference docs (platform tips, post templates, repurposing). There are no unexpected binaries, env vars, or install steps required by the registry metadata.
- Instruction Scope
- concernThe SKILL.md explicitly instructs the agent to read a workspace file (.claude/product-marketing-context.md) if it exists — that access was not declared in the skill metadata. The reverse-engineering reference also instructs collecting 500–1000+ posts and names scraping tools (Apify, PhantomBuster), which encourages large-scale data collection and may cause the agent to try to access external services or perform actions beyond simple content drafting. These instructions broaden the agent's operational scope beyond mere copywriting and should be noted.
- Install Mechanism
- okInstruction-only skill with no install spec and no code files. This minimizes risk from downloads or hidden executables.
- Credentials
- noteThe skill declares no credentials or config paths, but references tools and workflows (platform scrapers, export tools) that typically require API keys or third-party accounts. Also it asks to read a local file (.claude/product-marketing-context.md) which may contain sensitive product or business information. The absence of declared env vars is inconsistent with these possible needs.
- Persistence & Privilege
- okalways is false and there is no install behavior that modifies other skills or system-wide settings. The skill does not request permanent presence or elevated platform privileges.