unslop
Pass
Audited by Gen Agent Trust Hub on Nov 6, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill is composed entirely of markdown instructions for text editing. It does not contain any scripts, binary files, or network-enabled commands. The configuration explicitly disables tool invocation (
disable-model-invocation: true), which limits the agent's capability to take autonomous actions outside of text processing. - [INDIRECT_PROMPT_INJECTION]: The skill's primary function is to process user-provided text, which introduces a potential surface for indirect prompt injection. If the text submitted for editing contains hidden instructions, the agent could potentially be influenced to follow them.
- Ingestion points: Processes user-provided text strings for stylistic rewriting (SKILL.md).
- Boundary markers: The instructions do not define specific delimiters or "ignore instructions" wrappers for the text being processed.
- Capability inventory: The frontmatter configuration (
disable-model-invocation: true) prohibits the model from calling tools while this skill is active, mitigating the risk of tool-based exfiltration or privilege escalation. - Sanitization: No explicit sanitization or filtering is performed on the text input before it is processed by the agent.
Audit Metadata