Skillv1.0.0

ClawScan security

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

Scanner verdict

SuspiciousMar 11, 2026, 3:14 PM

Verdict: suspicious
Confidence: medium
Model: gpt-5-mini
Summary: The skill's files and instructions mostly match its business-growth purpose, but there are metadata and documentation inconsistencies (author/owner and stated count of skills) and the package includes runnable Python scripts that you should review for unexpected network or file-access behavior before installing.
Guidance: This package appears to be what it says (business growth analytics + templates) and uses local Python CLI tools, but there are several documentation/metadata mismatches that reduce trust. Before installing or running: 1) Verify the publisher/owner identity and repository referenced in the Quick Start (the manifest owner ID differs from the GitHub path/author in docs). 2) Inspect the included Python scripts for any network calls (urllib, http.client, socket), subprocess usage, or file reads outside the expected input (look for open(..., 'r') of absolute paths, os.environ access, or hardcoded endpoints/credentials). 3) Run the scripts in a sandbox or isolated environment on non-sensitive sample data first. 4) Search the code for strings like 'http', 'ftp', 'requests', 'socket', 'subprocess', 'Popen', 'open(', '~/', '/etc', or direct IP literals. 5) If you plan to use real customer data, get an explicit publisher provenance (GitHub repo, author contact) and consider a security review or automated static scan. These steps will reduce risk even if the package is benign.

Review Dimensions

Purpose & Capability: noteName/description match the included components: customer-success, sales-engineer, revenue-operations, and contract-and-proposal-writer are present and there are 9 Python scripts for scoring/analysis, which is coherent for a business-growth skillset. However, documentation inconsistencies reduce trustworthiness: SKILL.md / top-level description reference author 'Alireza Rezvani' and a GitHub Quick Start path 'alirezarezvani/claude-skills/business-growth' while registry metadata lists owner ID 'kn7f2gr00xy51fj1nx2y64ckjs800mhn' and source 'unknown'. CLAUDE.md also states '3/3 business & growth skills production-ready' while other files describe 4 skills. These mismatches don't prove malicious intent but are incoherent and worth validating with the publisher.
Instruction Scope: noteRuntime instructions are mostly local: run Python CLI tools against JSON input, use templates, and the docs explicitly state 'process files locally (no API calls)'. That scope is appropriate. The SKILL.md also embeds or exposes the full source and templates to review. Because executable Python scripts are included, the agent (or a user) could run arbitrary code if the scripts do network I/O or spawn subprocesses — the documentation claims stdlib-only and no API calls, but you should confirm the scripts don't access network, unexpected files, or environment variables before trusting them with real data.
Install Mechanism: okNo install spec is provided (instruction-only at install time) and there are no automated downloads or package installers described. That is a low-risk install mechanism. The Quick Start references an npx command and a GitHub path for an 'agent-skills-cli' add — that is a convenience instruction but not an automatic installer included in the package.
Credentials: okThe skill declares no required environment variables, no primary credential, and no config paths. That is proportionate for tools that operate on local JSON files and templates. Still confirm the included scripts do not read unexpected env vars or sensitive files at runtime (e.g., HOME, ~/.ssh, cloud credential files) despite the documentation claims.
Persistence & Privilege: okThe skill does not request always: true and is user-invocable only; model invocation is allowed (default) which is normal. There is no installation script that claims to modify other skills or system-wide settings. No persistent elevated privileges are requested.