Skill flagged — suspicious patterns detected
ClawHub Security flagged this skill as suspicious. Review the scan results before using.
GMGN Skill Token
v1.1.9On-chain token intelligence for Solana, BSC, and Base — security audit, rug pull risk, liquidity pool data, top holder distribution, and smart money / KOL tr...
⭐ 0· 60·0 current·0 all-time
by@gmgnai
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
OpenClaw
Suspicious
medium confidencePurpose & Capability
The skill claims to provide on-chain token intelligence via the gmgn-cli, which is coherent with the required binary. However, the metadata/requirements list no environment variables or config paths even though the SKILL.md clearly requires a GMGN_API_KEY and writes to ~/.config/gmgn/.env. This omission is an inconsistency between claimed requirements and actual instructions.
Instruction Scope
The SKILL.md instructs the agent/user to run gmgn-cli commands and to generate an Ed25519 keypair with openssl, then asks the user to paste an API key and stores it under ~/.config/gmgn/.env. The instructions leave a private key at /tmp/gmgn_private.pem (no secure-delete step) and direct writing secrets to disk. Those actions expand scope beyond merely querying token data and have privacy/security implications.
Install Mechanism
This is an instruction-only skill with no install spec; metadata suggests installing gmgn-cli from npm (npm install -g gmgn-cli). Installing from a public npm package is a reasonable, expected mechanism for a CLI tool; no arbitrary download URLs or extract steps are present.
Credentials
The skill requests no environment variables in its declared metadata but the runtime doc requires GMGN_API_KEY and writing to ~/.config/gmgn/.env. The missing declaration of required credentials/config paths is a notable mismatch. Asking users to paste an API key into the agent and storing it on disk increases risk if the agent or environment is untrusted.
Persistence & Privilege
always:false and model invocation are normal. The skill does instruct creating a user-scoped config (~/.config/gmgn/.env) to persist the API key and generates a temporary private key under /tmp. Persisting credentials in the user's home is expected for CLI tools but should be done carefully; the SKILL.md does not instruct secure handling/removal of the private key.
What to consider before installing
This skill appears to be a gmgn-cli wrapper, but it has some concerning omissions and secret-handling steps. Before installing or using it: (1) do not paste any API key unless you trust the skill and agent — if you must, prefer creating the API key directly in a trusted browser and store it yourself; (2) verify the source and authenticity of the gmgn-cli npm package (check the package owner, repository, and release history) before npm installing; (3) if you follow the SKILL.md setup, securely delete /tmp/gmgn_private.pem after generating the public key (e.g., shred or rm -P) and confirm the tool truly requires the keypair flow; (4) be aware the skill will write ~/.config/gmgn/.env containing your API key — consider restricting file permissions and reviewing that file before use; (5) ask the skill author/maintainer to update metadata to declare GMGN_API_KEY and the config path, and to provide explicit secure-delete and key-handling instructions. If the author cannot justify the missing metadata and the private-key instructions, treat this as untrusted and avoid sharing credentials with the agent.Like a lobster shell, security has layers — review code before you run it.
latestvk97a6r285rwymq2h9bkz6wby2x83ztq7
License
MIT-0
Free to use, modify, and redistribute. No attribution required.
Runtime requirements
🪙 Clawdis
Binsgmgn-cli