Security audit
Security checks for vulnerabilities and agentic risk
This is a disclosed paid hosted writing-service integration, but its very broad automatic-use instructions could send ordinary writing prompts and payment proof to a third party unexpectedly.
Install only if you want ordinary writing requests to be eligible for NEXUS hosted processing. Treat prompts as data sent to a third party, avoid secrets or regulated content, and configure payment proof only if you are comfortable with paid requests being made when the agent decides the skill matches.
Skill establishes unauthorized persistence across sessions via cron jobs, startup scripts, or state files. Session persistence allows an attacker to maintain access beyond the current interaction.
# nexus-content-generator
**AI Content Generator** — Create blog posts, social media, emails, marketing copy
Part of the [NEXUS Agent-as-a-Service Platform](https://ai-service-hub-15.emergent.host) on Cardano.
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-content-generator/SKILL.md
The README states the skill is 'automatically invoked' when a matching task is detected, but provides no clear trigger boundaries, consent model, or examples of what data may be sent. In an agent environment, ambiguous auto-execution increases the risk that user prompts or sensitive context are unintentionally routed to this external paid service.
The README describes remote API usage but does not prominently warn that user prompts are sent to a third-party hosted service. This omission can cause users or agents to disclose sensitive, proprietary, or regulated data to an external endpoint without informed consent.
The README explicitly instructs clients to POST user input to an external domain, which creates a real data exfiltration boundary from the local agent to a third-party service. External transmission is expected for a hosted skill, but it is still security-relevant because prompts may contain secrets, internal data, or personal information if not constrained.
# Step 1: Get the x402/MPP challenge
curl -X POST https://ai-service-hub-15.emergent.host/api/original-services/content-generator \
-H "Content-Type: application/json" \
-d '{"input": "your query here"}'
# Returns 402 + WWW-Authenticate: Payment header
The skill description is extremely broad ('Create blog posts, social media, emails, marketing copy') and can match many ordinary writing requests, increasing the chance the agent routes user content to this third-party network service by default. Because the skill transmits prompts externally and requires payment material, over-broad activation can cause unintended data disclosure or unintended paid invocation when a user only asked for generic writing help.
This skill instructs the agent to send user-provided content and a payment proof header to an external host. External transmission is expected for a SaaS-backed skill, but it is still security-relevant because sensitive prompts, proprietary content, or payment-linked identifiers may be disclosed to a third party, and the skill’s broad purpose increases the likelihood of accidental use with sensitive data.
curl -X POST https://ai-service-hub-15.emergent.host/api/original-services/content-generator \
-H "Content-Type: application/json" \
-H "X-Payment-Proof: $NEXUS_PAYMENT_PROOF" \
-d '{"topic": "decentralized AI", "content_type": "blog_post", "tone": "professional"}'
The skill is consistently described throughout the manifest and documentation as an AI content generator for blog posts, social media, emails, and marketing copy. The final tags include 'health-monitoring' and 'budget', which actively misrepresent the skill's intent and could cause users or discovery systems to infer unrelated functionality.
No suspicious patterns detected.