The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection as it performs actions and makes decisions based on the content of untrusted repository metadata and configuration files.
Ingestion points: The skill monitors and reads various configuration files such as pnpm-workspace.yaml, turbo.json, nx.json, lerna.json, package.json, Cargo.toml, go.work, and pyproject.toml to detect and manage monorepo structures.
Boundary markers: There are no explicit instructions or delimiters provided to the agent to treat the content of these external files as untrusted or to ignore any natural language instructions that might be embedded within them.
Capability inventory: The skill is granted access to the Read, Grep, Glob, and Bash tools, which allows it to read external data and subsequently execute shell commands based on the parsed information.
Sanitization: The instructions do not specify any validation or sanitization routines for data extracted from these configuration files before it is used to influence the agent's workflow or command execution.

monorepo