The Agent Skills Directory

[PERSISTENCE]: The scripts/install-statusline.py script modifies the settings.json file in the agent's configuration directory (typically ~/.claude/). It registers a statusLine command that points to the scripts/statusline.py renderer, ensuring the custom status line is active across agent sessions. This is the intended primary purpose of the skill.
[COMMAND_EXECUTION]: Several scripts execute local commands to perform their tasks. scripts/install-lefthook.mjs runs git and lefthook to manage git hooks. scripts/statusline.py may execute a previously configured status line command when operating in "composition" mode, sourcing the command from the user's existing settings stored in settings.json and a local state.json file.
[INDIRECT_PROMPT_INJECTION]: The skill's exploration phase involves reading various documentation and configuration files within the repository (e.g., AGENTS.md, .agents/notes/). This represents a potential surface for indirect prompt injection if those files were to contain malicious instructions. The skill mitigates this by instructing the agent to "Inspect files before editing or describing them; report only verified reading and completed work" and requiring explicit user confirmation before writing any changes.

setup-demon-skills