The Agent Skills Directory

[COMMAND_EXECUTION]: The skill is designed to execute a variety of shell commands to automate repository management. This includes Git operations (git worktree, git fetch, git remote), dependency management (npm install, pip install, composer install), and directory/file manipulation (cp, mkdir). It also attempts to launch external editors like VS Code.
[DATA_EXFILTRATION]: The instructions direct the agent to identify and copy .env files from the main project to new worktree locations. Although this is a local operation intended for environment parity, it represents an automated process that accesses and replicates sensitive configuration files.
[INDIRECT_PROMPT_INJECTION]: The skill analyzes external project files to determine the repository's configuration, which presents an attack surface for indirect instructions.
Ingestion points: The agent reads README.md, package.json, .gitignore, and sdd/ambiente.md from the project directory.
Boundary markers: There are no explicit markers or instructions used to delimit the content of these files or to prevent the agent from following instructions embedded within them.
Capability inventory: The skill has access to shell execution (git, npm), file system writes (sdd/ambiente.md), and directory creation (mkdir).
Sanitization: No sanitization or validation of the file content is performed before the information is used to drive agent behavior.
[EXTERNAL_DOWNLOADS]: The skill contains references to the author's GitHub repository for updates and provides commands to install additional tools using npx.

sdd-new-session