The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes local Protobuf files and project configuration files (such as buf.yaml, buf.gen.yaml, and buf.work.yaml) which act as ingestion points for external data that could contain malicious instructions.
Ingestion points: The skill reads .proto files and configuration metadata from the local repository as part of its standard workflow (Step 1 and Step 2).
Boundary markers: There are no explicit delimiters or specific instructions for the agent to ignore instructions that might be embedded within the comments or metadata of the files being analyzed.
Capability inventory: The skill is capable of executing shell commands via buf lint and performing file system write operations through buf format.
Sanitization: No explicit sanitization or validation of the input file content is performed before the tool processes the data or the agent interprets the results.

codeguard-protobuf