Skill flagged — suspicious patterns detected

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

Uplo Energy

v1.0.0

AI-powered energy sector knowledge management. Search power generation records, grid management data, regulatory filings, and safety protocols with structure...

0· 72·0 current·0 all-time
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
VirusTotalVirusTotal
Benign
View report →
OpenClawOpenClaw
Suspicious
medium confidence
Purpose & Capability
The skill claims to connect to an UPLO knowledge base and exposes search/graph tools — requiring an UPLO URL and API key is reasonable. However, the registry metadata lists no required env vars while skill.json declares config entries (agentdocs_url and api_key). That mismatch is incoherent: the skill will expect credentials even though the registry says none.
Instruction Scope
SKILL.md instructs the agent to run domain-specific tools (search_with_context, search_knowledge, get_directives, etc.) and to verify identity and classification before querying CEII. It does not instruct arbitrary file reads, unrelated credential collection, or exfiltration to unexpected endpoints beyond the UPLO instance.
!
Install Mechanism
There is no formal install spec in the registry, but skill.json defines an MCP runtime that uses `npx -y @agentdocs1/mcp-server --http`. That implies the runtime will fetch and run an npm package at use-time. Dynamic npx pulls are higher-risk than pre-reviewed installs because they execute remote code; confirm the @agentdocs1/mcp-server package provenance and contents before allowing the skill to invoke it.
Credentials
Requesting an UPLO URL and API key is proportional to a knowledge-base client. The concern is the missing declaration in registry-level required env vars and the skill manifest relying on those secrets. The skill will pass AGENTDOCS_URL/API_KEY to the MCP server process; ensure the API key has least privilege and that the host URL is trusted.
Persistence & Privilege
The skill does not request 'always: true' or any elevated persistent privileges. It appears to run on demand and does not modify other skills or system-wide settings in the provided materials.
What to consider before installing
This skill appears to be a client for an UPLO knowledge base and legitimately needs an UPLO URL and API key — but the package has three red flags you should resolve before installing: 1) Registry metadata vs manifest mismatch: the registry claims no required env vars, yet skill.json requires agentdocs_url and api_key. Treat the skill as requiring credentials until you verify otherwise. 2) Dynamic npm execution: the skill manifest runs `npx @agentdocs1/mcp-server`. That will fetch and execute a remote npm package at runtime. Verify the package name, its publisher, and inspect its source (or ask the vendor for a signed release) before allowing execution. 3) Unknown origin/homepage: the skill's source/homepage fields are empty even though README references uplo.ai. Verify the publisher identity (is this the official UPLO distribution?) and prefer skills with a clear source and release artifacts. Practical steps: - Confirm the publisher and check the npm package (@agentdocs1/mcp-server) source and recent versions. - Only provide an API key scoped with the minimum permissions and with an expiry/rotation policy, and point AGENTDOCS_URL to a trusted domain (prefer your organization's instance). - If possible, sandbox the skill’s MCP process or audit network activity when first running it. - If you need stronger assurance, ask the maintainer to update registry metadata to declare required env vars and provide a pinned, reviewable install artifact rather than implicit npx execution. If you cannot verify the package provenance or the publisher, treat this skill as higher-risk and do not provide sensitive credentials or CEII data to it.

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

latestvk9751z1x3jjyn2b87zew6y16m9839c4k

License

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

Comments