Skills
skills/full-statck-skills/utility-skills/codegraph/Gen Agent Trust Hub

codegraph

Fail

Audited by Gen Agent Trust Hub on Jun 16, 2026

Risk Level: CRITICALREMOTE_CODE_EXECUTIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [REMOTE_CODE_EXECUTION]: The skill's primary installation instructions in SKILL.md and references/installation.md direct users to download and execute a remote script using the command curl -fsSL https://raw.githubusercontent.com/colbymchenry/codegraph/main/install.sh | sh. Executing unverified remote scripts with shell privileges is a critical security risk.
  • [EXTERNAL_DOWNLOADS]: The tool relies on binary executables and Node.js packages (@colbymchenry/codegraph) hosted on external platforms (GitHub and NPM) owned by an untrusted third-party developer not recognized as a trusted organization.
  • [COMMAND_EXECUTION]: The skill utilizes several CLI commands such as codegraph install and codegraph init that modify system configuration and project files. These commands are executed via the unverified external CLI tool.
  • [PROMPT_INJECTION]: Instructions in references/troubleshooting.md and references/mcp-tools.md explicitly tell the AI agent to "trust the results" and "not use grep/read to repeat verification." This directive undermines the agent's ability to verify findings and increases the risk of successful indirect prompt injection from malicious project code.
Recommendations
  • HIGH: Downloads and executes remote code from: https://raw.githubusercontent.com/colbymchenry/codegraph/main/install.sh - DO NOT USE without thorough review
  • AI detected serious security threats
Audit Metadata
Risk Level
CRITICAL
Analyzed
Jun 16, 2026, 04:42 AM
Security Audit — agent-trust-hub — codegraph