Skill flagged — suspicious patterns detected

ClawHub Security flagged this skill as suspicious. Review the scan results before using.

Ai Specialists

v1.1.0

Interact with AI Specialists via the AI Specialists Hub MCP endpoint. Use when the user asks about any of their AI specialists (e.g. Ruby, Peter, Benjamin, M...

0· 617·1 current·1 all-time
byErik Ashby@erikashby
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
VirusTotalVirusTotal
Suspicious
View report →
OpenClawOpenClaw
Suspicious
medium confidence
!
Purpose & Capability
The skill claims no required binaries or environment variables, but includes a shell helper (scripts/mcp-call.sh) that clearly depends on curl and python3. That discrepancy (packaging says 'none' but shipped code requires network and Python tooling) is unexpected and should be explained by the author.
!
Instruction Scope
SKILL.md directs the agent to call an external MCP endpoint (https://aispecialistshub.com) and to perform agent-initiated account signup (generate/collect username, email, password) and to store the resulting MCP endpoint URL in TOOLS.md. Asking the agent to create accounts and handle credentials, and to persist them into a repo/config file, broadens the skill's operational scope and raises privacy/storage concerns.
Install Mechanism
This is an instruction-only skill (no install spec). Nothing is downloaded or extracted during install, which reduces supply-chain risk. The included script will be available but not auto-installed.
Credentials
The skill does not request environment variables or external credentials up front, which is consistent. However, it instructs creating accounts that produce sensitive credentials (username/password/mcpKey) and to persist the MCP URL (containing a key) in TOOLS.md — this is sensitive and should be treated as such. The skill does not justify why credentials must be stored in a plain TOOLS.md file.
Persistence & Privilege
always:false and normal invocation settings are appropriate. The skill asks the agent to write the MCP endpoint into TOOLS.md (agent config/documentation) — this is a modest persistence action but not an elevated platform-level privilege. Consider whether storing keys in TOOLS.md is acceptable in your environment.
What to consider before installing
This skill appears to do what it says (talk to an AI Specialists MCP endpoint) but has a few red flags you should consider before installing: - Verify the external service: aispecialistshub.com is the endpoint used for signup and MCP calls. Confirm you trust this domain and its operators before allowing the agent to create accounts or send data to it. - Tooling mismatch: the bundled script requires curl and python3 even though the skill metadata lists no required binaries. Ensure those binaries are present and review the script before running it. - Credentials handling: the skill supports agent-initiated signup and instructs the agent to generate/collect username, email, password, and an mcpKey, then store an MCP URL (containing the key) in TOOLS.md. Decide whether you want the agent creating accounts on your behalf and whether storing keys in TOOLS.md (likely plaintext) is acceptable. If not, require manual signup and manual injection of the MCP URL. - Least privilege: if you proceed, restrict network access where possible, require human approval before auto-signup, and avoid letting the agent persist credentials to public or shared repositories. If you want higher confidence, ask the skill author to (1) update metadata to declare required binaries (curl, python3), (2) document why credentials must be stored in TOOLS.md and offer a safer storage option, and (3) provide an audited privacy/security statement for aispecialistshub.com. If those items are addressed, the skill would be closer to benign.

Like a lobster shell, security has layers — review code before you run it.

latestvk97abw2w26b24wh3kd2cy5haq181jhqs

License

MIT-0
Free to use, modify, and redistribute. No attribution required.

Comments