Skills
skills/hotriluan/alkana-dashboard/gkg/Gen Agent Trust Hub

gkg

Fail

Audited by Gen Agent Trust Hub on Mar 29, 2026

Risk Level: HIGHREMOTE_CODE_EXECUTIONDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
  • [REMOTE_CODE_EXECUTION]: The installation process downloads and executes scripts directly from the official GitLab repository (gitlab.com/gitlab-org) using curl | bash for macOS/Linux and irm | iex for Windows systems.
  • [DATA_EXFILTRATION]: The skill starts a local HTTP server on port 27495 to facilitate API and MCP access. Documentation indicates that no authentication is required for local access, which may expose indexed codebase content to other processes running on the local machine.
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection as it enables the agent to retrieve and analyze raw source code through tools like read_definitions and get_definition.
  • Ingestion points: Project files and codebases indexed via the gkg index command.
  • Boundary markers: No specific delimiters or instructions are used to distinguish code content from instructions when data is passed to the agent.
  • Capability inventory: Includes indexing repositories, searching for symbols, and reading full definition bodies.
  • Sanitization: Source code content is provided to the agent without sanitization or filtering of embedded instructions or comments.
Recommendations
  • HIGH: Downloads and executes remote code from: https://gitlab.com/gitlab-org/rust/knowledge-graph/-/raw/main/install.sh - DO NOT USE without thorough review
Audit Metadata
Risk Level
HIGH
Analyzed
Mar 29, 2026, 03:17 AM