The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill fetches content from an external repository located at https://github.com/affaan-m/everything-claude-code.git using git clone to a temporary directory.
[COMMAND_EXECUTION]: The skill executes multiple shell commands to manage files and directories, including rm -rf, mkdir -p, and cp -r. These are used to prepare the environment and transfer downloaded components to the final installation targets.
[REMOTE_CODE_EXECUTION]: By copying downloaded .md files into the ~/.claude/skills/ directory, the skill facilitates the installation of new executable instructions. These files are subsequently loaded and executed by the AI agent as part of its capability set.
[PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface as it ingests untrusted data from a third-party repository and integrates it into the agent's configuration.
Ingestion points: Fetches data from https://github.com/affaan-m/everything-claude-code.git in SKILL.md.
Boundary markers: None present to separate or validate the integrity of downloaded skills/rules.
Capability inventory: Subprocess calls (git, rm, mkdir, cp, grep, ls) and file-write capabilities are used to install content into the agent's executable path.
Sanitization: No sanitization or verification of the remote content is performed before installation.

configure-ecc