Dada

• Outcome: Review
• Latest audit: Sep 12, 2026 · 2:07 AM UTC
• Version: 1.1.0

Overview

The skill is a delivery API guide, but it tells users to log all API traffic that can contain customer addresses, phone numbers, locations, and signatures.

Install only if you are comfortable treating it as developer guidance that needs privacy corrections. Do not follow the blanket logging recommendation in production; log only allowlisted operational fields, redact names, phone numbers, addresses, coordinates, callback URLs, signatures, and secrets, and use short retention with restricted log access.

A.I.G

Vulnerability Patterns

• Insecure Skill Coding Practices: Finds exploitable flaws such as hardcoded secrets or command injection
• Skill Instruction Hijacking: Alters the agent's session goals or safety constraints when the skill loads
• Agent Memory Poisoning: Writes attacker-controlled rules into memory that affect later sessions
• Remote Payload Retrieval and Execution: Fetches external code whose behavior can change after review
• Embedded Malicious Code: Ships malicious scripts inside the skill and executes them locally

Findings (1)

T09 · Insecure Skill Coding Practices

Warning

• Location: SKILL.md:243

Finding

Unrestricted API Traffic Logging May Expose Delivery Credentials and Personal Data

Content

View full analysis

Vulnerability Details

File Location: SKILL.md, lines 91–99 and line 243
Vulnerability Type: Sensitive data exposure through excessive logging
Risk Level: Medium

The Skill recommends recording all API requests and responses even though those messages can contain recipient names, telephone numbers, full addresses, precise coordinates, callback URLs, signatures, and other delivery information.

Relevant order fields from lines 91–99, translated into English:

text

| receiver_name    | String | Required | Recipient name |
| receiver_phone   | String | Required | Recipient phone number |
| receiver_address | String | Required | Recipient address |
| receiver_lat     | Double | Required | Recipient latitude |
| receiver_lng     | Double | Required | Recipient longitude |
| callback_url     | String | Required | Callback URL |

The vulnerable recommendation at line 243, translated into English:

text

3. Record all API requests and response logs to facilitate troubleshooting.

Technical Analysis

The recommendation uses an unrestricted logging policy rather than a minimum-data, allowlisted policy. API request bodies can contain recipient identity, contact information, full delivery addresses, and exact geographic coordinates. Responses and callbacks can also contain order details, courier names, courier phone numbers, identifiers, signatures, or operational metadata.

Logging entire request and response objects copies sensitive information from a comparatively short-lived transaction channel into logging systems, observability platforms, backups, support exports, and developer consoles. These secondary systems commonly have broader access and longer retention periods than the production API itself.

The behavior is not required for the Skill's declared delivery-integration functionality. Troubleshooting can instead rely on request IDs, sanitized error codes, endpoint names, status codes, and ...[truncated 1823 chars]

Remediation

View remediation

Remediation Suggestions

1. Replace the instruction to log all traffic with a structured, allowlisted logging policy.
2. Log only operationally necessary fields, such as:
   • Internal request or correlation ID
   • API endpoint name
   • HTTP status code
   • Sanitized provider error code
   • Request duration
   • Retry count
3. Never log complete request or response bodies in production by default.
4. Redact or omit names, phone numbers, full addresses, coordinates, callback tokens, signatures, application secrets, and other authentication material.
5. If temporary payload diagnostics are unavoidable, require explicit authorization, use synthetic or masked data, and automatically disable verbose logging after a short period.
6. Apply least-privilege access controls to logging and observability systems.
7. Encrypt logs in transit and at rest, enforce short retention periods, and prevent sensitive records from being copied into unrestricted support exports or backups.
8. Add automated log-scrubbing tests to verify that sensitive fields cannot enter production logs.
9. Document an incident-response process for rotating credentials and deleting retained records if sensitive API traffic is accidentally logged.

SkillSpector

Vulnerability Patterns

• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Privilege Escalation: Excessive Permissions, Sudo/Root Execution, Credential Access
• Supply Chain: Unpinned Dependencies, External Script Fetching, Obfuscated Code
• Excessive Agency: Unrestricted Tool Access, Autonomous Decision Making, Scope Creep

Findings (1)

Missing User Warnings

Medium

• Category: Not specified by scanner
• Confidence: 94% confidence

Finding

The skill includes example request payloads and field definitions containing personal data such as receiver names, phone numbers, addresses, geolocation, and courier contact details, but it provides no explicit guidance on privacy, minimization, masking, retention, or secure handling. In an integration-focused skill, developers may copy these patterns directly into logs, test fixtures, or prompts, increasing the risk of PII exposure through insecure storage, over-collection, or disclosure.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.