The Agent Skills Directory

[EXTERNAL_DOWNLOADS] (LOW): The skill utilizes pnpm nx add and uv add to install external packages. While these are unverifiable dependencies, their installation is a primary function of monorepo management, warranting a severity downgrade.
[DATA_EXFILTRATION] (LOW): The command npx nx connect facilitates a connection to Nx Cloud (nx.app) for remote caching and build telemetry. This involves network operations to a non-whitelisted domain.
[COMMAND_EXECUTION] (LOW): The skill relies on executing shell commands for task orchestration (nx run-many), code generation (nx g), and environment synchronization (uv sync).
[PROMPT_INJECTION] (LOW): As a development tool, the skill is vulnerable to indirect prompt injection through the analysis of workspace files.
Ingestion points: The skill reads nx.json, project.json, pyproject.toml, and source code within the monorepo.
Boundary markers: Absent; there are no explicit instructions for the agent to treat workspace data as untrusted or to ignore embedded instructions.
Capability inventory: The agent can execute arbitrary shell commands via the Nx CLI and modify the filesystem through generators.
Sanitization: Absent; configuration data from the workspace is passed directly to CLI tools.

nx-monorepo