prompt-compression
Pass
Audited by Gen Agent Trust Hub on Sep 29, 2026
Risk Level: SAFENO_CODEINDIRECT_PROMPT_INJECTION
Full Analysis
- [NO_CODE]: The skill contains only Markdown instructions and YAML metadata. There are no scripts, binaries, or automated installation steps included in the package.
- [INDIRECT_PROMPT_INJECTION]: The skill processes external prompt text provided by the user. While this constitutes an ingestion surface for indirect prompt injection, the skill does not have any dangerous capabilities (e.g., network access, file system modifications, or shell execution) that could be leveraged by malicious instructions.
- Ingestion points: Processes user-provided prompts to be compressed (SKILL.md).
- Boundary markers: The instructions do not specify any markers or delimiters to isolate the prompt content being processed.
- Capability inventory: No tool access or system capabilities are available to the skill.
- Sanitization: No sanitization is performed on the input text as the skill's purpose is a semantic transformation of that text.
Audit Metadata