Security audit
Security checks for vulnerabilities and agentic risk
This skill is a clearly disclosed paid remote data-validation API wrapper, with no local file or shell authority, but users should treat submitted input as leaving their environment.
Install only if you are comfortable sending the task input to NEXUS for server-side AI processing and paying the disclosed per-request fee when not using the sandbox. Do not submit passwords, private keys, personal data, regulated records, or confidential business data unless your organization has approved this provider and payment flow.
Skill enumerates or reads other installed skills. Access to other skills' SKILL.md files or the skills directory reveals prompt instructions, capabilities, and secrets that should be invisible to peer skills.
Or manually copy the SKILL.md to your OpenClaw skills directory:
cp SKILL.md ~/.openclaw/skills/nexus-data-validation/SKILL.md
The README states the skill is 'automatically invoked' when a matching task is detected, but provides no concrete trigger boundaries, scope limits, or consent expectations. In an agent environment, vague auto-invocation language can lead to unintended activation on sensitive tasks and unexpected transmission of user data to the external service.
The README describes sending request data to a third-party remote API and using a payment-bearing authorization flow, but it does not clearly warn users that their inputs leave the local agent boundary. This creates privacy, compliance, and financial risk because users or downstream agents may unknowingly transmit sensitive data and incur charges.
The documented curl example explicitly posts user-provided input to an external host, which is a real data egress behavior. In the context of an auto-invoked agent skill, this is dangerous if sensitive prompts, proprietary datasets, or regulated information are transmitted without clear disclosure, minimization, and consent controls.
# Step 1: Get the x402/MPP challenge
curl -X POST https://ai-service-hub-15.emergent.host/api/original-services/data-validation \
-H "Content-Type: application/json" \
-d '{"input": "your query here"}'
# Returns 402 + WWW-Authenticate: Payment header
The usage guidance says to use the skill for 'intelligent data quality checks and schema validation' without defining data sensitivity boundaries, allowed data classes, or safe-use constraints. Because the skill forwards user input to a third-party remote AI service, vague invocation guidance increases the chance that agents or users will send secrets, regulated data, or internal records off-platform without informed consent.
The skill explicitly instructs sending arbitrary input and a payment proof header to an external endpoint controlled by a third party. Even though this is the advertised purpose of the skill, it creates a real data-exfiltration path: anything passed as input, including potentially sensitive data, is transmitted off the local agent boundary, and the trust/privacy statements are only assertions by the provider rather than enforceable controls.
curl -X POST https://ai-service-hub-15.emergent.host/api/original-services/data-validation \
-H "Content-Type: application/json" \
-H "X-Payment-Proof: $NEXUS_PAYMENT_PROOF" \
-d '{