Skills
Navigate
skills/ses2905/ai-toolkit/install-work-kit/Gen Agent Trust Hub

install-work-kit

Pass

Audited by Gen Agent Trust Hub on Sep 20, 2026

Risk Level: SAFEREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [REMOTE_CODE_EXECUTION]: The skill provides instructions to clone the author's repository from GitHub and execute local shell scripts to complete the installation process.
  • Evidence: The skill instructs the agent to run git clone https://github.com/Ses2905/cursor-skills.git followed by ./scripts/install-local.sh and ./scripts/sync-user-skills.sh.
  • [COMMAND_EXECUTION]: Utilizes the npx tool to fetch and integrate additional skill content from the remote repository.
  • Evidence: npx skills add https://github.com/Ses2905/cursor-skills.
  • [INDIRECT_PROMPT_INJECTION]: The skill serves as a mechanism for installing a large number of additional agent skills and plugins, which expands the set of instructions and data the agent will process in future sessions.
  • Ingestion points: The entire work-kit package is ingested from an external repository during the installation flow in SKILL.md.
  • Boundary markers: None identified in the installation instructions.
  • Capability inventory: Execution of local shell scripts and file system writes to the ~/.cursor/plugins/local/ directory.
  • Sanitization: Not present for the installation commands.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 20, 2026, 05:58 AM