Skillv1.1.0

ClawScan security

NEXUS Commit Message · ClawHub's context-aware review of the artifact, metadata, and declared behavior.

BenignApr 9, 2026, 5:08 PM

Verdict: benign
Confidence: high
Model: gpt-5-mini
Summary: This skill is an instruction-only wrapper for a paid external commit-message API; its requirements and instructions are consistent with that purpose, but it sends your diffs to a third-party service and expects a payment proof credential.
Guidance: This skill sends your diffs to a third-party service (https://ai-service-hub-15.emergent.host) and requires a payment-proof credential. Only install if you trust that host. Consider these before installing: (1) Do you accept sending repository diffs (possibly sensitive code) to an external provider? (2) Store NEXUS_PAYMENT_PROOF only in a secure place (avoid exposing it in public or checked-in env files). (3) Test with sandbox_test first to confirm behavior and cost. (4) Verify the service's documentation and TLS certificate, and prefer local/offline tools if you cannot share code with an external paid API.
Findings: [no_regex_findings] expected: The scanner found nothing—expected because this is an instruction-only skill with no code files to analyze. Absence of findings is not a guarantee of safety; the SKILL.md instructs network calls to an external endpoint.

Purpose & Capability: okThe skill's name/description (generate commit messages from diffs) matches the SKILL.md, which directs calls to an external NEXUS API. Requesting a payment-proof credential (NEXUS_PAYMENT_PROOF) is coherent with a paid remote service.
Instruction Scope: noteRuntime instructions only send the provided input to the external endpoint and explain payment flows (x402 / MPP / legacy). The instructions explicitly require network access and state there is no filesystem or shell access. This stays inside the stated purpose, but it does involve transmitting user input (diffs) to a third-party service—users should expect that.
Install Mechanism: okNo install spec and no code files are included (instruction-only). Nothing is written to disk by the skill itself, which minimizes install risk.
Credentials: noteOnly a single env var (NEXUS_PAYMENT_PROOF) is required, which is proportionate for a paid API. However, this value is sensitive (payment credential/proof). The SKILL.md allows a sandbox value ('sandbox_test') for testing, so requiring a persistent env var may be inconvenient but not incoherent.
Persistence & Privilege: okThe skill is not always-enabled and does not request elevated persistence or modify other skills. Autonomous invocation is allowed by default (normal).