Skillv1.0.0

ClawScan security

Ads URL Parser · ClawHub's context-aware review of the artifact, metadata, and declared behavior.

BenignMar 4, 2026, 5:58 AM

Verdict: benign
Confidence: high
Model: gpt-5-mini
Summary: The skill is an instruction-only URL-to-ads-brief parser and its declared inputs and behavior are coherent with its stated purpose; it does not request unrelated credentials or install code.
Guidance: This skill appears internally consistent and safe to review, but keep these practical cautions in mind: 1) The agent will need to fetch whatever URL you provide — do not submit private/internal URLs, URLs containing tokens, or pages with sensitive data you don't want transmitted. 2) Parsing remote pages can expose the agent to malicious or obfuscated content (scripts, trackers); verify how your agent fetches and sanitizes HTML. 3) The skill does not request ad platform credentials; if you later ask the agent to perform account-level actions it may prompt for them — provide credentials only to trusted, intended integrations. If you need higher assurance, ask the skill owner for a description of the fetch/sanitization behavior or run it in an environment that mediates outbound web requests.

Purpose & Capability: okName, description, input/output contracts, and decision rules align: the skill's purpose is to parse URLs and produce ad launch briefs, and it does not request unexpected binaries, env vars, or installs.
Instruction Scope: noteSKILL.md stays within the ad-briefing scope (validate URL, extract offer/CTA, map KPIs, emit JSON). It implicitly requires fetching/parsing the provided URL (network access and HTML parsing), but it does not instruct reading unrelated system files, environment variables, or sending data to third-party endpoints beyond the user-supplied URL. Recommend confirming how your agent will fetch/handle remote content (follow redirects, execute JS, sanitize scripts) before using with sensitive links.
Install Mechanism: okNo install spec or code files; instruction-only skills have low install risk since nothing is written to disk or fetched during install.
Credentials: okNo credentials, env vars, or config paths are required. The declared inputs (URL, business_goal, market_scope) are proportionate to the stated functionality.
Persistence & Privilege: okalways is false and the skill does not request persistent system-level privileges or modify other skills' configuration. Autonomous invocation is allowed by default but not combined with other red flags.