ClawHub
Back to skill

Security audit

AI Agent Trading on DEX

Security checks for vulnerabilities and agentic risk

Overview

This skill is a clearly disclosed trading API helper, but users should treat it as high-impact because it can manage live trading deployments with real funds.

Install only if you understand that the Superior Trade API key can create and manage deployments that may trade real funds. Keep the key scoped and stored securely, review every live deployment summary before approving it, and confirm deletion or cleanup actions because they can close positions and orders.

A.I.G

Vulnerability Patterns
  • 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
  • Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
  • Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
  • Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs

SkillSpector

Vulnerability Patterns
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Tool MisuseTool Parameter Abuse, Chaining Abuse, Unsafe Defaults
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (10)

Tool Parameter Abuse

High
Category
Tool Misuse
Confidence
80% confidence
Finding

Tool parameters are crafted to achieve unintended or unsafe behavior. Parameter abuse can bypass intended safety checks (e.g. shell=True, --force, dangerous glob patterns).

Content

Scanner excerpt · SKILL.md (reported line 251)May include surrounding context.

md
5. `GET /v2/backtesting/{id}` — fetch full results; download `result_url` for detailed JSON
6. Present summary: total trades, win rate, profit, drawdown, Sharpe ratio
7. If failed, check `GET /v2/backtesting/{id}/logs`
8. To cancel: `DELETE /v2/backtesting/{id}`

#### Result Interpretation

Tool Parameter Abuse

High
Category
Tool Misuse
Confidence
80% confidence
Finding

Tool parameters are crafted to achieve unintended or unsafe behavior. Parameter abuse can bypass intended safety checks (e.g. shell=True, --force, dangerous glob patterns).

Content

Scanner excerpt · SKILL.md (reported line 691)May include surrounding context.

md
5. `GET /v2/backtesting/{id}` — fetch full results; download `result_url` for detailed JSON
6. Present summary: total trades, win rate, profit, drawdown, Sharpe ratio
7. If failed, check `GET /v2/backtesting/{id}/logs`
8. To cancel: `DELETE /v2/backtesting/{id}`

#### Result Interpretation

Tool Parameter Abuse

High
Category
Tool Misuse
Confidence
80% confidence
Finding

Tool parameters are crafted to achieve unintended or unsafe behavior. Parameter abuse can bypass intended safety checks (e.g. shell=True, --force, dangerous glob patterns).

Content

Scanner excerpt · SKILL.md (reported line 277)May include surrounding context.

md
a. Show the user a table: name, status, created date for each deployment
   b. Ask: "You have [n]/10 deployments. Which stopped/unused ones can I delete to make room?"
   c. Wait for explicit selection or approval
   d. `DELETE /v2/deployment/{id}` for each approved — note: this closes all positions/orders first
3. Only proceed to create once count is **9 or fewer** (leaving room for the new one)

If count is already **10**, do NOT attempt `POST /v2/deployment` — it will fail. Show the list and require cleanup first.

Tool Parameter Abuse

High
Category
Tool Misuse
Confidence
80% confidence
Finding

Tool parameters are crafted to achieve unintended or unsafe behavior. Parameter abuse can bypass intended safety checks (e.g. shell=True, --force, dangerous glob patterns).

Content

Scanner excerpt · SKILL.md (reported line 684)May include surrounding context.

md
a. Show the user a table: name, status, created date for each deployment
   b. Ask: "You have [n]/10 deployments. Which stopped/unused ones can I delete to make room?"
   c. Wait for explicit selection or approval
   d. `DELETE /v2/deployment/{id}` for each approved — note: this closes all positions/orders first
3. Only proceed to create once count is **9 or fewer** (leaving room for the new one)

If count is already **10**, do NOT attempt `POST /v2/deployment` — it will fail. Show the list and require cleanup first.

External Transmission

Medium
Category
Data Exfiltration
Confidence
50% 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 30)May include surrounding context.

md
externalEndpoints:
  - url: https://api.superior.trade
    purpose: "All backtesting and deployment operations"
  - url: https://api.hyperliquid.xyz/info
    purpose: "Read-only public queries. Balance checks send the user's public wallet address (not a secret — visible on-chain). Pair validation sends no user data. No authentication or secrets are sent to this endpoint."
---

External Transmission

Medium
Category
Data Exfiltration
Confidence
50% 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 147)May include surrounding context.

md
externalEndpoints:
  - url: https://api.superior.trade
    purpose: "All backtesting and deployment operations"
  - url: https://api.hyperliquid.xyz/info
    purpose: "Read-only public queries. Balance checks send the user's public wallet address (not a secret — visible on-chain). Pair validation sends no user data. No authentication or secrets are sent to this endpoint."
---

External Transmission

Medium
Category
Data Exfiltration
Confidence
50% 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 229)May include surrounding context.

md
externalEndpoints:
  - url: https://api.superior.trade
    purpose: "All backtesting and deployment operations"
  - url: https://api.hyperliquid.xyz/info
    purpose: "Read-only public queries. Balance checks send the user's public wallet address (not a secret — visible on-chain). Pair validation sends no user data. No authentication or secrets are sent to this endpoint."
---

External Transmission

Medium
Category
Data Exfiltration
Confidence
50% 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 230)May include surrounding context.

md
externalEndpoints:
  - url: https://api.superior.trade
    purpose: "All backtesting and deployment operations"
  - url: https://api.hyperliquid.xyz/info
    purpose: "Read-only public queries. Balance checks send the user's public wallet address (not a secret — visible on-chain). Pair validation sends no user data. No authentication or secrets are sent to this endpoint."
---

External Transmission

Medium
Category
Data Exfiltration
Confidence
50% 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 304)May include surrounding context.

md
externalEndpoints:
  - url: https://api.superior.trade
    purpose: "All backtesting and deployment operations"
  - url: https://api.hyperliquid.xyz/info
    purpose: "Read-only public queries. Balance checks send the user's public wallet address (not a secret — visible on-chain). Pair validation sends no user data. No authentication or secrets are sent to this endpoint."
---

Behavior Manipulation

Medium
Category
Prompt Injection
Confidence
75% confidence
Finding

Subtle instructions detected that may alter agent decision-making or introduce hidden biases.

Content

Scanner excerpt · SKILL.md (reported line 133)May include surrounding context.

md
- The credentials endpoint returns `wallet_type: "agent_wallet"` for auto-resolved wallets
- Always check the **main wallet's** balance, not the agent wallet's
- The API has no transfer/fund-routing endpoint — you cannot move funds via the API
- **NEVER tell users to deposit to the agent wallet address**

### Funding and Balance Checks

Static analysis

No suspicious patterns detected.