Security audit
Security checks for vulnerabilities and agentic risk
This skill is a disclosed paid network bridge to a third-party NEXUS API, with no artifact evidence of hidden local file access, shell execution, persistence, or deception.
Install only if you trust NEXUS with the prompts you send and with payment proof verification. Do not include secrets, private repository contents, or sensitive filesystem data in requests unless that is intentional, and prefer sandbox or tightly scoped payment credentials for testing.
The skill is described as a bridge to IPFS, GitHub, and the filesystem but provides no warning that prompts, files, repository data, or other sensitive content may be transmitted or that local system state may be affected. In this context, missing privacy and system-impact disclosures materially increase the risk of users enabling a powerful connector without informed consent.
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-mcp-bridge/SKILL.md
The README states the skill is 'automatically invoked' for 'matching tasks' without defining the trigger scope, data shared, or approval boundaries. For a bridge that can connect to external MCP servers including GitHub and the filesystem, vague activation criteria can cause unintended invocation and unexpected data exposure or side effects.
The documented usage sends arbitrary user input to an external hosted service, which is an external transmission channel. In a bridge skill context, this is expected behavior, but it is still security-relevant because sensitive prompts or file-derived content could be sent off-device to a third party without clear minimization or warning.
# Step 1: Get the x402/MPP challenge
curl -X POST https://ai-service-hub-15.emergent.host/api/original-services/mcp-bridge \
-H "Content-Type: application/json" \
-d '{"input": "your query here"}'
# Returns 402 + WWW-Authenticate: Payment header
The skill explicitly instructs the agent to transmit user-provided input and a payment proof token to an external third-party endpoint. Even though this is the advertised purpose of the skill, it still creates a real data-exfiltration and secret-exposure risk because the request sends potentially sensitive prompts off-platform and includes an environment-derived credential in a header.
curl -X POST https://ai-service-hub-15.emergent.host/api/original-services/mcp-bridge \
-H "Content-Type: application/json" \
-H "X-Payment-Proof: $NEXUS_PAYMENT_PROOF" \
-d '{
No suspicious patterns detected.