praise-before-work
Pass
Audited by Gen Agent Trust Hub on Oct 18, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill's primary function is to modify the agent's tone and response structure. Analysis of the instructional content and configuration files found no evidence of malicious intent or security vulnerabilities.
- [EXTERNAL_DOWNLOADS]: The README documentation references an installation method using the vendor's own CLI tool (
lovstudio). As this is a resource owned by the skill's author, it is categorized as a standard deployment procedure and not a security risk. - [INDIRECT_PROMPT_INJECTION]: The skill is designed to process user-supplied tasks to generate relevant praise. The instructions include safety guardrails that direct the agent to remain honest (not acknowledging incorrect conclusions to please the user) and to respect specific formatting constraints (like "results only" requests), which mitigates common risks associated with processing untrusted input.
- Ingestion points: User tasks and greetings processed at runtime in
SKILL.md. - Boundary markers: None explicitly defined in the prompt template, but behavioral instructions serve as logical boundaries.
- Capability inventory: Text generation and task execution as delegated by the host agent.
- Sanitization: None detected beyond logical instructions to maintain accuracy.
Audit Metadata