Back to skill
Skillv1.0.0
ClawScan security
neur0pay · ClawHub's context-aware review of the artifact, metadata, and declared behavior.
Scanner verdict
SuspiciousApr 29, 2026, 2:27 PM
- Verdict
- suspicious
- Confidence
- medium
- Model
- gpt-5-mini
- Summary
- The skill's instructions match its stated NeuroPay purpose but contain manifest/instruction inconsistencies (missing declared env var, contradictory guidance about asking the user) and the publisher is unknown — review before installing.
- Guidance
- What to check before installing: - Confirm the publisher and site: the skill has no homepage and an opaque owner ID. Verify neuropay.fr is the legitimate API you intend to use and the skill author is trustworthy. - Manifest mismatch: SKILL.md requires NEUROPAY_API_KEY but the skill metadata did not declare any required env vars or primary credential — ask the publisher to fix the manifest to explicitly require NEUROPAY_API_KEY. - Clarify agent behavior: the instructions contradict themselves about whether to prompt the user for the key. Decide whether you want the agent to prompt, and only provide the key via a secure secret store or environment variable if you trust the skill. - Limit the API key: if you proceed, use a scoped key with minimal permissions and rotate it afterward. Don’t reuse a high-privilege or long-lived key. - Network exposure: the skill will make authenticated requests to https://neuropay.fr; ensure that is expected and acceptable for your environment. - If you need higher assurance, request the publisher provide a verifiable repo/homepage, update SKILL.yaml to declare the NEUROPAY_API_KEY, and add documentation about key handling and missing-key behavior. If you cannot verify the publisher or resolve the manifest/instruction contradictions, consider not installing or running this skill with real credentials.
Review Dimensions
- Purpose & Capability
- concernThe SKILL.md clearly requires a NeuroPay API key and documents API endpoints on neuropay.fr which is coherent with the stated purpose. However, the skill metadata declares no required environment variables or primary credential — that is inconsistent. Also there is no homepage and owner is an opaque ID, reducing traceability.
- Instruction Scope
- noteThe runtime instructions direct only to call neuropay.fr API endpoints and to include the API key header; they do not instruct reading unrelated system files or other secrets. However there are internal contradictions: the agent is told to 'Automatically use' NEUROPAY_API_KEY and 'Never ask the user', yet also says 'If missing: Please provide your NeuroPay API key'. The instructions also admonish never to store or log the key (good), but the contradictory behavior guidance creates ambiguity about how the agent should behave when the key is absent.
- Install Mechanism
- okNo install spec and no code files — instruction-only skill. This minimizes disk/execute risk since nothing is being downloaded or installed by the skill package itself.
- Credentials
- concernThe skill uses a secret NEUROPAY_API_KEY according to SKILL.md, which is a reasonable and proportionate single credential for this API. But the manifest does not declare this required environment variable or primary credential. That mismatch could lead to confusion or accidental user input handling. The skill does not request other unrelated credentials.
- Persistence & Privilege
- okalways is false, user-invocable and autonomous invocation are default. The skill does not declare any config paths or request persistent system-level presence. No indication it modifies other skills or global settings.