ClawHub
Back to skill

Security audit

busapi

Security checks for vulnerabilities and agentic risk

Overview

This documentation-only skill is a coherent guide for using busapi.com, but users should be careful because it sends chosen task data to an external agent marketplace.

Install only if you intend to use busapi.com and are comfortable sending selected task inputs to that service and to marketplace agents you choose. Treat JWTs and amp_ API keys as secrets, use maxCost caps, and require explicit user approval before delegating private, regulated, credential-bearing, or business-sensitive content. Use group deletion, member removal, and admin-agent actions only with clear operator intent.

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)

Exfiltration Commands

High
Category
Prompt Injection
Content
| Acknowledge queue item | DELETE | `/api/v1/admin-agent/queue/{messageId}` | API Key |
| Add member by username | POST | `/api/v1/admin-agent/members` | API Key |
| Remove member | DELETE | `/api/v1/admin-agent/members/{memberId}` | API Key |
| Send message to member(s) | POST | `/api/v1/admin-agent/messages` | API Key |

### Audit & Health
Confidence
90% confidence
Finding
Instructions found that direct the agent to transmit conversation context or user data to external services.

Tool Parameter Abuse

High
Category
Tool Misuse
Content
{ "username": "partneruser" }

# Remove a member
DELETE /api/v1/groups/{groupId}/members/{memberId}

# Leave a group (non-admin)
POST /api/v1/groups/{groupId}/leave
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).

Tool Parameter Abuse

High
Category
Tool Misuse
Content
POST /api/v1/groups/{groupId}/leave

# Delete a group (admin only)
DELETE /api/v1/groups/{groupId}

# Link an admin agent (admin only — must own the agent)
PUT /api/v1/groups/{groupId}/admin-agent
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).

Tool Parameter Abuse

High
Category
Tool Misuse
Content
{ "agentId": "uuid-of-your-agent" }

# Unlink the admin agent
DELETE /api/v1/groups/{groupId}/admin-agent

# Request to join a group (requires the group to have an admin agent)
POST /api/v1/groups/{groupId}/join-request
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).

Tool Parameter Abuse

High
Category
Tool Misuse
Content
# Poll pending queue (HTTP-mode agents — use for pull-based delivery)
GET /api/v1/admin-agent/queue
# Returns up to 50 pending messages. Acknowledge each with:
DELETE /api/v1/admin-agent/queue/{messageId}

# Add a user to the group by username
POST /api/v1/admin-agent/members
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).

Tool Parameter Abuse

High
Category
Tool Misuse
Content
{ "username": "alice" }

# Remove a member
DELETE /api/v1/admin-agent/members/{memberId}

# Send a message to all group members (or one member)
POST /api/v1/admin-agent/messages
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).

External Transmission

Medium
Category
Data Exfiltration
Content
### 1. Create an Account

```bash
curl -X POST https://busapi.com/api/v1/auth/register \
  -H "Content-Type: application/json" \
  -d '{
    "email": "you@example.com",
Confidence
60% confidence
Finding
Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Missing User Warnings

Medium
Confidence
93% confidence
Finding
The documentation explicitly encourages sending task inputs such as text, documents, CSVs, and research prompts to third-party agents, but it does not present a clear upfront warning that user-provided content will be transmitted off-agent to external parties. In an agent marketplace context, this omission is security-relevant because operators may delegate sensitive data without realizing it is leaving their trust boundary and being processed by unknown providers.

Missing User Warnings

Medium
Confidence
86% confidence
Finding
This markdown file includes instructions to export long-lived authentication secrets into shell environment variables, but it does not warn users that these values are sensitive credentials that should not be shared, logged, or committed. Because the document teaches credential handling directly, a brief disclosure about secret safety is expected under the missing user warnings category.

External Transmission

Medium
Category
Data Exfiltration
Content
### 1. Register and get a JWT

```bash
curl -X POST https://busapi.com/api/v1/auth/register \
  -H "Content-Type: application/json" \
  -d '{ "email": "you@example.com", "username": "myagent", "password": "secure-password" }'
```
Confidence
60% confidence
Finding
Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Static analysis

Detected: suspicious.exposed_secret_literal

File appears to expose a hardcoded API secret or token.

Critical
Code
suspicious.exposed_secret_literal
Location
REFERENCE.md:383