Skill flagged — suspicious patterns detected
ClawHub Security flagged this skill as suspicious. Review the scan results before using.
Keyapi Pinterest Analysis
v1.0.0Discover and analyze Pinterest users, pins, boards, followers, and following — search users, retrieve profile information, explore pin libraries and board co...
⭐ 0· 28·0 current·0 all-time
by@lycici
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
OpenClaw
Benign
high confidencePurpose & Capability
Name/description (Pinterest analysis) match the actual requirements and behavior: the skill needs node and a KEYAPI_TOKEN and calls a KeyAPI MCP server. There are no unrelated credentials or binaries requested.
Instruction Scope
SKILL.md instructs running scripts/run.js which lists tools, reads schemas, calls MCP tools, caches responses, and can auto-paginate. This stays within the stated purpose, but the tool persists API responses to a local cache and can save the API token to a .env file (the script will write .env). Also the client supports overriding the MCP endpoint via KEYAPI_SERVER_URL, which could redirect calls to an arbitrary server if misconfigured.
Install Mechanism
No download/install script from arbitrary URLs. It uses npm dependencies (package.json) — specifically @modelcontextprotocol/sdk — which is a normal package registry install. No extract-from-URL or non-standard installers were used.
Credentials
Only KEYAPI_TOKEN is declared as required and that is appropriate for an API-backed integration. The code also reads an optional KEYAPI_SERVER_URL (not listed in required.env) to override the server; this is legitimate but worth noting because it can redirect traffic to another endpoint if set.
Persistence & Privilege
always is false and the skill does not request elevated platform privileges. The script writes a .keyapi-cache directory and may write a .env file containing KEYAPI_TOKEN in the skill directory, persisting credentials and API responses locally. The skill can be invoked autonomously by agents (default), which increases blast radius if a token is present.
Assessment
This skill is coherent with its stated Pinterest-analysis purpose, but review the following before installing: 1) The skill will persist API responses into .keyapi-cache and may save your KEYAPI_TOKEN to a .env file in the skill directory — treat that token like a secret. 2) The code supports overriding the MCP endpoint via KEYAPI_SERVER_URL; ensure this is not set to an untrusted URL. 3) Installing runs npm install which will fetch @modelcontextprotocol/sdk from the npm registry — review that dependency if you need extra assurance. 4) Because the agent can invoke skills autonomously by default, avoid leaving a high-privilege token in environment variables if you do not want automated requests made. If any of these behaviors are unacceptable, do not install or remove the token/override before running.scripts/run.js:52
Environment variable access combined with network send.
scripts/run.js:37
File read combined with network send (possible exfiltration).
Patterns worth reviewing
These patterns may indicate risky behavior. Check the VirusTotal and OpenClaw results above for context-aware analysis before installing.Like a lobster shell, security has layers — review code before you run it.
latestvk97awy3gp6sbd68dyv806t0rmn843f2a
License
MIT-0
Free to use, modify, and redistribute. No attribution required.
Runtime requirements
📌 Clawdis
Binsnode
EnvKEYAPI_TOKEN
Primary envKEYAPI_TOKEN