Back to skill
Skillv1.0.2
ClawScan security
Ecom Seller Profit · ClawHub's context-aware review of the artifact, metadata, and declared behavior.
Scanner verdict
SuspiciousApr 29, 2026, 1:01 AM
- Verdict
- suspicious
- Confidence
- medium
- Model
- gpt-5-mini
- Summary
- The skill's core functionality (parsing order exports and calculating profit) appears consistent with its name, but there are important mismatches and missing declarations — notably undeclared runtime dependencies and required billing credentials that will be sent to an external billing endpoint — that you should understand before installing.
- Guidance
- This skill appears to implement the advertised profit-calculation functionality, but several transparency issues should be resolved before you run it. Key points to consider: - Metadata mismatch: The skill registry entry claims no required env vars or binaries, yet SKILL.md and parse_orders.py require SKILL_BILLING_API_KEY, SKILL_BILLING_SKILL_ID, and FEISHU_USER_ID and the scripts need Python + openpyxl. Ask the publisher to update the registry metadata to list these required environment variables and runtime dependencies. - Billing and data sharing: The skill will call an external billing endpoint (skillpay.me) and transmit FEISHU_USER_ID for billing. Decide whether you are comfortable your user ID will be sent to that third party. Ask for exact request payloads and privacy details (what else is transmitted? is any order content uploaded for billing?) If you don't want this, do not provide those env vars or do not install the skill. - Secrets handling: SKILL_BILLING_API_KEY is sensitive. Only provide it if you trust the billing provider and the skill owner. Prefer using credentials issued specifically for this skill with limited scope, and rotate/revoke them after testing. - Dependency safety: Because there is no install spec, verify the runtime environment has Python and openpyxl. Avoid running the script in an environment with high privileges or with access to unrelated credentials. If you must test, run it in an isolated container or VM with sample data. - Source and provenance: The skill's source/homepage are unknown. Ask the publisher for a repository or signed release so you can audit the full parse_orders.py (the provided file appears truncated in the registry listing) and confirm exactly what network calls are made and what is sent. If the publisher can correct the registry metadata to list required env vars and dependencies, provide clear documentation of the billing request payloads, and/or provide vetted hosting/source, the concerns become much smaller. Until then treat the skill as untrusted for sensitive data and test in isolation.
Review Dimensions
- Purpose & Capability
- concernThe skill's name, SKILL.md, and Python scripts all align on the stated purpose (analyzing order exports and calculating per-order/platform profit). However the registry metadata claims no required environment variables or binaries, while the package contains Python scripts that declare/expect billing environment variables and require Python + openpyxl. The metadata omission is a coherence problem: a paid skill that invokes an external billing service should declare those env vars and runtime requirements in the registry entry.
- Instruction Scope
- concernSKILL.md and the scripts instruct the agent to read Excel/CSV order exports locally (analyze_headers.py, parse_orders.py) and perform LLM-based header mapping. In addition the skill performs per-execution billing via SkillPay (skillpay.me/api/v1/billing/charge) and explicitly transmits FEISHU_USER_ID to SkillPay for billing. Transmitting a user identifier to an external service is outside pure parsing functionality and is sensitive — it's explained in the docs but not reflected in the registry metadata. No other unrelated files or system paths are requested, and analyze_headers.py does not call external services.
- Install Mechanism
- noteThere is no install spec (instruction-only), so nothing is automatically downloaded. However the included Python scripts require Python and the openpyxl library; those dependencies are not declared in the registry metadata or SKILL.md's 'Required binaries' section. That means the skill may fail at runtime or require the agent to install packages manually, which is a usability and transparency issue (and raises risk if an agent tries to auto-install missing packages).
- Credentials
- concernSKILL.md and parse_orders.py list three required environment variables for billing: SKILL_BILLING_API_KEY, SKILL_BILLING_SKILL_ID, and FEISHU_USER_ID. These are plausible for a paid skill, but the registry metadata lists no required env vars — an inconsistency. SKILL_BILLING_API_KEY is a sensitive secret, and FEISHU_USER_ID is a user identifier explicitly sent to an external billing endpoint (skillpay.me). Requesting an API key and sending a user-id is proportionate to billing but must be declared transparently in the registry metadata; the omission is a red flag.
- Persistence & Privilege
- okThe skill does not request always:true, does not modify other skills, and does not declare persistent or system-wide privileges. It runs as an instruction-run script and appears to store nothing beyond optional output files (JSON/Markdown).