lov-install-web-logo
Pass
Audited by Gen Agent Trust Hub on Oct 18, 2026
Risk Level: SAFECOMMAND_EXECUTIONPERSISTENCEINDIRECT_PROMPT_INJECTIONDATA_EXFILTRATION
Full Analysis
- [COMMAND_EXECUTION]: The skill README and instructions reference local commands for setup (npx), validation (scripts/validate_skill.py), and profile management (scripts/profile_store.py). These are standard operational tasks for the skill's intended development environment.
- [PERSISTENCE]: The skill utilizes a specialized script (profile_store.py) to maintain user identity, brand assets, and workspace preferences in standard local configuration directories (~/.config/agent-skills/). This behavior is part of the skill's core functionality to persist state across sessions.
- [INDIRECT_PROMPT_INJECTION]: The skill processes project manifests, HTML templates, and component source code. This represents a potential attack surface if local project files contain malicious instructions, although the risk is mitigated by the skill's specific focus on branding metadata and image assets.
- [DATA_EXFILTRATION]: The skill accesses local configuration files to manage user and brand profiles. The provided scripts include explicit checks to prevent the storage of sensitive data such as tokens or passwords in these local records.
Audit Metadata