skills/kensaurus/skills/enhance-pwa/Gen Agent Trust Hub

enhance-pwa

Pass

Audited by Gen Agent Trust Hub on Oct 7, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted project data by auditing local configuration files (such as package.json, vite.config., and next.config.) to determine implementation strategies. This behavior creates an attack surface where an attacker-controlled project could attempt to influence the agent's behavior during the enhancement process. \n
  • Ingestion points: Audits local project files in SKILL.md Phase 0. \n
  • Boundary markers: The instructions do not specify the use of delimiters or directives to ignore embedded instructions within the audited source files. \n
  • Capability inventory: The skill performs file system write operations for manifest and service worker generation and suggests executing build-time CLI tools like sharp. \n
  • Sanitization: There is no mention of sanitizing or validating the content of the project files before they are processed by the agent. \n- [SAFE]: The skill implements established web security best practices, explicitly warning the agent and user never to cache authentication tokens or sensitive API responses. The dependencies and tools recommended are well-known and standard within the web development community.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 7, 2026, 11:27 PM