The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: Fetches the Moru Core library from the author's GitHub repository at https://github.com/taehwandev/moru-core and retrieves the vibeguard package from the npm registry.
[REMOTE_CODE_EXECUTION]: Executes Python scripts from the downloaded Moru Core library (e.g., workflow.py, setup-agent-hooks.py) to perform project validation and setup.
[REMOTE_CODE_EXECUTION]: Executes the @taehwandev/vibeguard package using npx, which involves downloading and running code from the npm registry at runtime.
[COMMAND_EXECUTION]: Runs shell commands such as vibeguard audit, vibeguard setup, and vibeguard update to manage project guardrails and instructions.
[PERSISTENCE]: Modifies AI agent runtime configurations in the user's home directory, including ~/.claude/settings.json, ~/.gemini/config/hooks.json, and ~/.antigravity/AGENTS.md, to install persistent lifecycle hooks and routing pointers.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process documentation files from the target repository, creating a potential surface for indirect prompt injection.
Ingestion points: The agent is instructed to read various repository files such as AGENTS.md, CLAUDE.md, and CONTRIBUTING.md to understand the project's current instructions.
Capability inventory: The skill has the capability to write to the filesystem, modify agent configurations, and execute shell commands and scripts.
Boundary markers: There are no explicit instructions to use delimiters or ignore embedded instructions when reading these project files.
Sanitization: No specific sanitization or filtering logic is prescribed for the content retrieved from the target repository documentation.

agent-bootstrap