Skill flagged — suspicious patterns detected
ClawHub Security flagged this skill as suspicious. Review the scan results before using.
Uplo Engineering
v1.0.0AI-powered engineering knowledge management. Search architecture docs, API specs, incident reports, runbooks, and infrastructure documentation with structure...
⭐ 0· 77·0 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
The skill's functionality (searching org docs via a UPLO MCP server) matches its name and description. However the registry metadata claims 'no required env vars / credentials', while skill.json and README clearly require a UPLO instance URL and an API key. That mismatch is incoherent and could mislead users about what secrets the skill needs.
Instruction Scope
SKILL.md instructs the agent to call MCP tools (search_with_context, search_knowledge, export_org_context, get_directives, log_conversation, etc.). These are in-scope for an engineering knowledge tool, but they will cause the agent to transmit queries and potentially sensitive document contents to the configured UPLO instance. The instructions do not attempt to read unrelated local files or other system secrets, but they do direct the agent to surface identity/context (get_identity_context) which may include access/clearance info.
Install Mechanism
Although the package is instruction-only in the registry summary, skill.json and the README expect the platform to run an MCP server via 'npx -y @agentdocs1/mcp-server --http'. That means code will be pulled from npm and executed at runtime (npx) — a moderate-risk install pattern. The npm package '@agentdocs1/mcp-server' is not a known, audited vendor here; downloading and running it automatically is a potential supply-chain risk and should be reviewed before use.
Credentials
The skill legitimately needs an 'agentdocs_url' and an 'api_key' to contact your UPLO instance and access your org's documents; this is proportionate to the purpose. The concern is the earlier registry metadata that reported no required credentials — that inconsistency is misleading. Also, the api_key will grant access to potentially very sensitive org data, so it should be scoped and protected (least privilege).
Persistence & Privilege
The skill does not request persistent 'always' inclusion and does not declare system-wide config changes. Its MCP server runs as a tool process (mcp block) and autonomous model invocation is allowed by default (not flagged here). There is no indication the skill modifies other skills or system-wide settings.
What to consider before installing
Before installing: 1) Note the registry metadata is inconsistent — the skill actually requires a UPLO instance URL and an API key (check skill.json/README). 2) The skill will run an MCP server via 'npx @agentdocs1/mcp-server' which downloads and executes code from npm — review that package's source and trustworthiness before allowing it to run. 3) The provided API key will allow the skill to read and log org documentation and incident reports — ensure the key is scoped with least privilege, audited, and used in a staging environment first. 4) Confirm privacy/retention policies for the UPLO instance (what gets logged/sent back to the vendor). 5) If you cannot verify the npm package or vendor, request a self-hostable or audited binary, or ask the publisher for a homepage/source repo and provenance. 6) If you proceed, start with a limited-access test account and monitor network/process activity and UPLO logs.Like a lobster shell, security has layers — review code before you run it.
latestvk976m2kb2c8bvvtemh60zesa3d838szy
License
MIT-0
Free to use, modify, and redistribute. No attribution required.