Back to skill

Security audit

OneABC

Security checks for vulnerabilities and agentic risk

Overview

This skill is a disclosed API wrapper, but it can send an API key and prompts to a configurable endpoint, including via an automatic OPENAI_API_KEY fallback.

Review before installing. Use a dedicated OneABC-scoped API key, avoid relying on OPENAI_API_KEY fallback, and only set ONEABC_BASE_URL to an endpoint you fully trust. Do not send sensitive prompts unless you are comfortable with them being transmitted to that configured API service.

Vulnerability Patterns
  • Insecure Skill Coding PracticesFinds exploitable flaws such as hardcoded secrets or command injection
  • 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
Findings (1)

T09 · Insecure Skill Coding Practices

Warning
Location
scripts/oneabc.js:3
Finding

API credentials and user prompts can be transmitted to an unrestricted configurable endpoint

Content
View full analysis
m.id).join('\n')); break; } ``` ### Technical Analysis The script selects `ONEABC_API_KEY` but automatically falls back to `OPENAI_API_KEY`. It then places the selected credential directly in the `Authorization` header of requests sent to the destination controlled by `ONEABC_BASE_URL`. The request to `/v1/models` is necessary for the declared model-listing functionality when it uses a genuine OneABC credential and the trusted OneABC endpoint. Likewise, prompt transmission is intrinsic to the declared generation features. However, two implementation choices weaken the trust boundary: 1. A credential associated with another provider can be reused automatically. 2. The destination receiving that credential and user content is unrestricted and controlled through an environment variable. C ...[truncated 2371 chars]
Remediation
View remediation
` rather than an unqualified credential. ]]>
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (3)

Undeclared Tool Scope

Medium
Category
MCP Least Privilege
Confidence
91% confidence
Finding

The skill invokes a local Node script that depends on environment variables and performs network access, but the manifest does not declare any explicit tool scope such as permissions or allowed-tools. This creates an authorization and review gap: operators cannot clearly see or constrain that the skill can read secrets from the environment and send data to an external API endpoint.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
92% confidence
Finding

The script sends the user-provided prompt to a remote API and includes an authorization credential in the request headers, but there is no confirmation prompt, user-facing notice, or explanatory comment/docstring disclosing that data will be transmitted off-system. For a general-purpose CLI helper, this is a safety-relevant operation that should be made explicit to users.

Content

No source excerpt is available for this finding.

Missing User Warnings

Low
Category
Not specified by scanner
Confidence
84% confidence
Finding

The script reads ONEABC_API_KEY or OPENAI_API_KEY from the environment, which are sensitive credentials, but only reports when the key is missing. It does not inform the user that the skill depends on and consumes these credentials.

Content

No source excerpt is available for this finding.