Skills
▥
skills/inference-sh-9/skills/technical-blog-writing/Gen Agent Trust Hub

technical-blog-writing

Fail

Audited by Gen Agent Trust Hub on Feb 25, 2026

Risk Level: HIGHREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [REMOTE_CODE_EXECUTION]: The documentation provides an installation command curl -fsSL https://cli.inference.sh | sh. This is a vendor-provided script for setting up the necessary CLI environment and originates from the author's official domain.
  • [COMMAND_EXECUTION]: The skill uses the infsh CLI to run various applications, such as infsh/python-executor for data visualization and infsh/html-to-image for generating blog assets. These are controlled executions of vendor-signed tools.
  • [EXTERNAL_DOWNLOADS]: The skill references downloading binaries and checksums from dist.inference.sh as part of its setup instructions, which is a standard distribution method for this vendor.
  • [PROMPT_INJECTION]: The skill exhibits an attack surface for indirect prompt injection as it processes external data from exa/search to inform blog content. • Ingestion points: Data from exa/search is ingested to research topics. • Boundary markers: No delimiters are specified to isolate untrusted search content from instructions. • Capability inventory: The skill has access to tools that execute code, generate files, and interact with external APIs (X/Twitter). • Sanitization: There is no evidence of content filtering or sanitization of search results.
Recommendations
  • HIGH: Downloads and executes remote code from: https://cli.inference.sh - DO NOT USE without thorough review
Audit Metadata
Risk Level
HIGH
Analyzed
Feb 25, 2026, 05:36 PM