vp-tldr
Pass
Audited by Gen Agent Trust Hub on Oct 7, 2026
Risk Level: SAFENO_CODEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: No malicious patterns or security risks were identified in the instruction set or configuration files.
- [NO_CODE]: The skill consists entirely of descriptive Markdown files and a YAML configuration defining an AI agent's writing style; it contains no scripts, binaries, or automated tasks.
- [INDIRECT_PROMPT_INJECTION]: The skill identifies a potential attack surface by processing untrusted user-supplied content to generate summaries. While it lacks explicit sanitization or delimiters to prevent the agent from being influenced by instructions embedded within the source text, it also has no dangerous capabilities (network access, file writes, or command execution) that could be exploited. Ingestion: Reads user content to summarize (SKILL.md). Boundaries: Absent. Capabilities: None. Sanitization: Absent.
Audit Metadata