Skill flagged — suspicious patterns detected
ClawHub Security flagged this skill as suspicious. Review the scan results before using.
Uplo Education
v1.0.0AI-powered education knowledge management. Search curriculum documents, student records frameworks, accreditation data, and institutional research with struc...
⭐ 0· 127·0 current·0 all-time
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
OpenClaw
Suspicious
medium confidencePurpose & Capability
Name, README, SKILL.md, and skill.json all describe an UPLO connector for institutional education knowledge (search, GraphRAG, exports). The required config (agentdocs_url and api_key in skill.json) is coherent with a connector to an external UPLO/agentdocs service.
Instruction Scope
SKILL.md instructs the agent to call domain-specific tools (search_knowledge, search_with_context, export_org_context, etc.) and to respect data classification and FERPA. It does not instruct reading unrelated system files, scanning local creds, or exfiltrating data to unexpected endpoints within the provided text.
Install Mechanism
There is no explicit install spec in the registry record, but skill.json config instructs launching an MCP server via `npx @agentdocs1/mcp-server --http`. That will download and execute a scoped npm package at runtime. Using npx to fetch and run code is a legitimate connector pattern but is higher risk than an instruction-only skill because it pulls third‑party code from the npm registry; the package scope (@agentdocs1) is not a widely-known public project in this manifest, so you should verify the package's origin and contents before enabling.
Credentials
skill.json declares two required configs: agentdocs_url and api_key (an MCP token). Those are proportionate for a connector to an external UPLO instance. However, the top-level 'Requirements' section in the registry summary stated 'Required env vars: none' which contradicts skill.json. This metadata inconsistency could lead to confusion during install or inadvertent misconfiguration.
Persistence & Privilege
The skill is not always-enabled and does not request special system paths or persistent privileges. It uses the normal MCP model invocation flow; autonomous invocation is permitted (platform default) and appropriate for this connector.
What to consider before installing
This skill appears to be a legitimate connector to an UPLO/AgentDocs knowledge service, but take these precautions before installing:
- Verify the source: ask the publisher for a homepage or repository and review the @agentdocs1/mcp-server npm package code (or vendor-supplied binaries) before allowing the agent to run npx to fetch it.
- Resolve the metadata inconsistency: the registry summary says no env vars are required but skill.json requires agentdocs_url and api_key. Confirm you will provide a properly scoped API key and correct instance URL.
- Use least-privilege credentials: create an API key scoped only to the data and operations this skill needs (read/search, limited write if propose_update/flag_outdated are enabled). Ensure audit logging is enabled on the UPLO instance.
- Run first in an isolated/test environment: because npx will download and execute remote code, enable the skill initially in a sandboxed agent or with network egress controls.
- Confirm FERPA/data handling expectations: the skill will surface institutional/student data—make sure that classification tiers, access controls, and data retention policies meet your compliance requirements.
If the vendor cannot point you to a verifiable package source or explain why the registry metadata omits the required creds, do not enable the skill until you have that information.Like a lobster shell, security has layers — review code before you run it.
latestvk97fh9gvqsy4z070mvt6vh2mw18383bk
License
MIT-0
Free to use, modify, and redistribute. No attribution required.