Skill flagged — suspicious patterns detected
ClawHub Security flagged this skill as suspicious. Review the scan results before using.
x402-payment-tron
v0.0.4Pay for x402-enabled Agent endpoints using USDT on TRON
⭐ 2· 3.4k·1 current·3 all-time
byopen-aibank@hades-ye
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
OpenClaw
Suspicious
medium confidencePurpose & Capability
Name/description (x402 payments on TRON) align with code and dependencies (tronweb, @open-aibank/x402-tron). However registry metadata lists no required env vars while SKILL.md and the code do expect TRON_PRIVATE_KEY — a mismatch between declared requirements and actual needs.
Instruction Scope
SKILL.md and the bundled code instruct the tool to locate and load a private key from multiple places: TRON_PRIVATE_KEY env var, x402-config.json and ~/.x402-config.json, and ~/.mcporter/mcporter.json (it will search inside mcpServers entries). The SKILL.md also contains an 'Agent Note' telling agents not to search for keys, which contradicts the tool's own behavior. The tool also performs an 'infinite approval' (MAX_UINT256) for USDT allowance — expected for a payments tool but high-risk behavior that should be explicitly consented to by users.
Install Mechanism
No external download/install spec is provided; the skill is instruction-only but includes source and a prebuilt dist bundle. Dependencies are standard npm modules (tronweb and an x402 client). No suspicious remote download URLs or extract steps were found in the manifest.
Credentials
Requesting a TRON private key is proportionate for a payments skill, but the skill quietly reads additional local configuration files (including ~/.mcporter/mcporter.json) and will look across server entries for TRON_PRIVATE_KEY — this broad file access is more intrusive than declaring a single env var. Registry metadata failing to declare the primary credential increases the risk of accidental exposure.
Persistence & Privilege
The skill does not request always:true, does not modify other skills, and does not persist beyond its own runtime. It writes temporary files for binary/image responses to the system temp directory and asks the agent to delete them — a standard pattern but worth cleaning up after use.
Scan Findings in Context
[base64-block] expected: The code decodes a base64 PAYMENT-RESPONSE header to parse settlement info; a base64 pattern was detected in SKILL.md but this usage is expected for payment metadata and not necessarily malicious.
What to consider before installing
This skill appears to do what it says (make USDT-on-TRON payments), but there are several concerning inconsistencies you should address before installing:
- The package actually needs a TRON private key, but the registry metadata does not list any required env vars. Assume you must provide TRON_PRIVATE_KEY and verify that in your environment settings.
- The bundled code will silently search local files for keys (x402-config.json, ~/.x402-config.json, and ~/.mcporter/mcporter.json) and may extract TRON_PRIVATE_KEY from them. If you have sensitive configs in those locations, consider moving them or using a dedicated wallet for this skill. Prefer setting TRON_PRIVATE_KEY explicitly in a secure environment variable rather than relying on file discovery.
- The tool performs an "infinite approval" (MAX_UINT256) for USDT allowance if needed. That reduces friction but is high risk: if the contract or underlying keys are compromised, funds could be drained. Only use with wallets that hold limited funds and consider manually approving only needed amounts.
- There is an internal contradiction: the SKILL.md tells the agent not to search for keys, but the tool itself does. Treat the tool as the authoritative behavior and review the code yourself or request the author to remove silent file scanning.
- If you decide to proceed, audit the included dist bundle or run the skill in an isolated environment (sandbox or throwaway VM), use a low-value wallet, and monitor/revoke token approvals (revoke infinite approval) after use.
If you need higher confidence, ask the publisher to (1) update registry metadata to declare TRON_PRIVATE_KEY, (2) remove silent scanning of home config files or make it opt-in, and (3) offer an explicit UX/confirmation before broadcasting infinite approvals.Like a lobster shell, security has layers — review code before you run it.
latestvk97etzhg4ga5mpafvh0qx0a6k580e3v1
License
MIT-0
Free to use, modify, and redistribute. No attribution required.
Runtime requirements
💳 Clawdis
