linkedin-poll-generator
Pass
Audited by Gen Agent Trust Hub on Sep 29, 2026
Risk Level: SAFE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill reads and follows instructions from voice configuration files (
.liftli/voice.mdand~/.liftli/voice.md) to align content with the user's persona. This ingestion of external data is a standard feature for personalizing content. - Ingestion points: Project-level and user-level voice configuration files (
voice.md) described in SKILL.md. - Boundary markers: None; the skill explicitly instructs the agent to follow the content of the file.
- Capability inventory: Limited to text generation for polls and potential interaction with publishing tools.
- Sanitization: No validation or sanitization of the voice file content is specified.
- [EXTERNAL_DOWNLOADS]: The skill documentation suggests that users install additional components and MCP servers from the author's official domains (
liftli-aiandliftli.ai). These references are consistent with standard vendor ecosystem extensions.
Audit Metadata