ClawHub
Back to skill

Security audit

Maiat Trust API

Security checks for vulnerabilities and agentic risk

Overview

This skill is mostly aligned with agent trust workflows, but it needs Review because it links wallet identities, wraps wallet transactions with a third-party package, and sends reputation/threat data to an external service.

Review before installing in any wallet-bearing or production agent. Use a dedicated wallet or pseudonymous identity, pin and audit the npm packages before use, confirm what transaction and report data Maiat receives, and avoid enabling automatic blocking/reporting unless its privacy, failure, and appeal behavior match your risk tolerance.

A.I.G

Vulnerability Patterns
  • Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs
  • Tool Hijacking and SpoofingModifies or replaces tools so legitimate-looking calls execute attacker logic
  • Insecure DependenciesIntroduces malicious components through unsafe dependency sources
  • Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
  • Agent Memory PoisoningWrites attacker-controlled rules into memory that affect later sessions
Findings (3)

T08 · Insecure Dependencies

Warning
Location
SKILL.md:167
Finding

Unpinned Third-Party Dependencies Create a Supply-Chain Risk

Content
View full analysis
Remediation
View remediation

T07 · Tool Hijacking and Spoofing

Warning
Location
SKILL.md:171
Finding

Third-Party Wrapper Intercepts Wallet Transactions and Performs External Reporting

Content
View full analysis
Remediation
View remediation

T05 · Unauthorized Access and Privilege Escalation

Warning
Location
SKILL.md:31
Finding

Unnecessary External Registration and Persistent Wallet-Identity Linking

Content
View full analysis
Remediation
View remediation

SkillSpector

Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (9)

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

The skill instructs users to register an on-chain identity by submitting an ENS-style name and wallet address to an external service without a prominent privacy warning. Because the action creates persistent public linkage and external disclosure of identity data, users may unintentionally deanonymize agents or operators and expose long-lived attribution metadata.

Content

No source excerpt is available for this finding.

External Transmission

Medium
Category
Data Exfiltration
Confidence
91% confidence
Finding

This is a real external transmission point because the example sends ENS and wallet identity data to a remote API for registration. In the context of a trust/identity skill, such transmission is expected, but it remains security-relevant because it shares sensitive identifiers with a third party and may establish persistent public identity linkage.

Content

Scanner excerpt · SKILL.md (reported line 36)May include surrounding context.

Every agent needs a Maiat Passport — your on-chain identity.

bash
curl -X POST https://app.maiat.io/api/v1/passport/register \
  -H "Content-Type: application/json" \
  -d '{"ensName": "my-agent", "walletAddress": "0xYourWallet", "type": "agent"}'

External Transmission

Medium
Category
Data Exfiltration
Confidence
60% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Content

Scanner excerpt · SKILL.md (reported line 151)May include surrounding context.

Lookup Any Passport

bash
curl https://app.maiat.io/api/v1/passport/lookup?q=agent-name

Deep Agent Profile

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The Guard section promotes automatic transaction blocking and threat reporting but does not prominently disclose that transaction intent, counterparties, and threat telemetry may be sent to an external service for evaluation. In a wallet-protection context this is sensitive operational data, and silent blocking/reporting can surprise users and affect financial operations or privacy.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The skill strongly encourages always reporting outcomes with a wallet identifier, but does not warn that this sends job/activity metadata tied to the reporter address to an external service. This can create a detailed behavioral trail linking agent activity, counterparties, and success/failure outcomes to a persistent identity.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

Medium
Category
Not specified by scanner
Confidence
91% confidence
Finding

The manifest describes the skill as trust infrastructure for registering agent identities, checking agent trust scores, and protecting transactions between agents. However, the file also exposes substantial token-analysis functionality, including honeypot detection, rug-pull prediction, liquidity analysis, and multi-source token forensics, which goes beyond verifying another agent before transacting.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
92% confidence
Finding

The threat-reporting flow allows user-submitted reports to contribute to ecosystem-wide blocking after three reports, yet it does not clearly warn about the downstream consequences or external sharing of submitted intelligence. Without strong disclosure and safeguards, users may unknowingly participate in actions that affect third parties at network scale and expose sensitive investigative data.

Content

No source excerpt is available for this finding.

External Transmission

Medium
Category
Data Exfiltration
Confidence
90% confidence
Finding

This is a true external transmission because it sends threat intelligence about an address to a third-party API, and the resulting reports can influence blocking decisions across the ecosystem. The sensitivity is increased by the combination of external disclosure, possible reputational harm to third parties, and insufficient warning about network-wide effects.

Content

Scanner excerpt · SKILL.md (reported line 333)May include surrounding context.

Report malicious addresses to protect the entire network:

bash
curl -X POST https://app.maiat.io/api/v1/threat/report \
  -H "Content-Type: application/json" \
  -d '{"maliciousAddress": "0x...", "threatType": "address_poisoning"}'

Description-Behavior Mismatch

Low
Category
Not specified by scanner
Confidence
80% confidence
Finding

The manifest frames the skill around identity, trust-score lookup, and transaction protection, but the documented functionality also includes outcome reporting, review submission, scarab balance checks, threat reporting, and agent reputation services. These features materially expand the skill from a verifier/protector into a broader reputation and feedback network.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.