ClawHub
Back to skill

Security audit

api-designer

Security checks for vulnerabilities and agentic risk

Overview

This is an API design helper made of Markdown instructions and examples; its credential-related examples are relevant to the purpose but should be reviewed before production use.

Reasonable to install as an API design assistant. Do not paste production secrets into prompts, and treat generated SDK authentication code as a starting point: add retry limits, token-refresh preconditions, idempotency protections for state-changing requests, and secure secret storage before production use.

SkillSpector

Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (2)

Context-Inappropriate Capability

Medium
Confidence
83% confidence
Finding
The generated client class includes live authentication header handling and automatic retry behavior after 401 responses, including a placeholder token refresh flow. This is dangerous because consumers may copy the pattern into production with incomplete refresh logic, causing insecure token handling, unintended request replay, or retry loops against authenticated endpoints.

Missing User Warnings

Medium
Confidence
79% confidence
Finding
The example silently injects API keys and bearer tokens into outbound HTTP headers and configures a client for network transmission without any warning about secret handling. In a skill framed as API design, this is risky because users may not expect executable credential-using client behavior and may adopt insecure patterns for secret propagation and logging.

Static analysis

Detected: suspicious.exposed_secret_literal

File appears to expose a hardcoded API secret or token.

Critical
Code
suspicious.exposed_secret_literal
Location
references/examples.md:21