The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to run package manager commands such as pnpm install and build commands like pnpm turbo build after modifying configuration files like package.json across the workspace.
[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it parses content from repository files that could be controlled by an external actor. If a user opens a malicious repository, configuration files like package.json could contain strings designed to influence the agent's behavior.
Ingestion points: Configuration files including package.json, pnpm-workspace.yaml, turbo.json, nx.json, and lerna.json in the working directory.
Boundary markers: The skill does not define specific delimiters or instructions to ignore embedded prompts when reading these files.
Capability inventory: Includes reading and writing filesystem files and executing shell commands (package installations and builds).
Sanitization: There is no evidence of sanitization or validation of the data extracted from the configuration files before it is processed by the agent.

monorepo-manager