configure-ecc
Warn
Audited by Gen Agent Trust Hub on Oct 14, 2026
Risk Level: MEDIUMEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: Clones repository content from the public GitHub repository at
affaan-m/everything-claude-codeinto a temporary directory for installation. - [COMMAND_EXECUTION]: Uses several shell commands to manage files and directories, including
git clone,mkdir -p,cp -r(to install files into sensitive agent directories like~/.claude/skills/),ls -la,grep, andrm -rf. - [INDIRECT_PROMPT_INJECTION]: The skill facilitates the installation of third-party instructions and rules into the agent's environment. As these materials originate from an unverified external source, they represent a potential attack surface where malicious prompts or logic could be introduced and subsequently executed by the agent in future sessions. The analysis identifies a combined read-and-write capability involving untrusted remote data and persistent agent configuration files.
- [DYNAMIC_EXECUTION]: Provides functionality to 'optimize' the installed skills by reading and editing the
SKILL.mdand rule files in-place at the installation target. This involves the agent programmatically modifying its own instructional logic at runtime based on user tech stack requirements.
Audit Metadata