Skills
▥
skills/inference-sh-9/skills/newsletter-curation/Gen Agent Trust Hub

newsletter-curation

Fail

Audited by Gen Agent Trust Hub on Feb 25, 2026

Risk Level: HIGHEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: Fetches the infsh CLI tool from the vendor's official domain (cli.inference.sh).- [REMOTE_CODE_EXECUTION]: Includes a command to download and execute a shell script from the vendor's domain (curl -fsSL https://cli.inference.sh | sh). This is a standard installation pattern for the vendor's utility.- [COMMAND_EXECUTION]: Executes shell commands via the infsh tool to interact with search assistants and image generation apps.- [PROMPT_INJECTION]: The skill possesses an indirect prompt injection surface by ingesting untrusted data from external sources.
  • Ingestion points: Results from tavily/search-assistant and exa/search tools are used to curate content.
  • Boundary markers: No explicit delimiters or instructions are provided to the agent to distinguish between untrusted search results and its core instructions.
  • Capability inventory: The skill has the capability to post content externally via the x/post-create tool.
  • Sanitization: No sanitization or validation of the external search results is described.
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