The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill provides instructions for the agent to process Protobuf (.proto) files and configuration files. This data ingestion surface is susceptible to indirect prompt injection, where malicious instructions could be hidden in comments or metadata within the Protobuf files to influence the agent when it reads the tool's output.
Ingestion points: Protobuf (.proto) files, buf.yaml, and buf.gen.yaml as described in SKILL.md and references/ci-integration.md.
Boundary markers: The instructions do not specify the use of delimiters or provide warnings to the agent to ignore embedded instructions within the files being processed.
Capability inventory: The skill facilitates shell command execution and interacts with the file system via the buf CLI.
Sanitization: No sanitization or filtering of the content within the processed files is mentioned before the agent interacts with the tool output.
[COMMAND_EXECUTION]: The skill describes the use of shell commands to install and run the buf CLI tool (e.g., buf build, buf lint, buf breaking). These commands allow the agent to perform file system operations and compilation tasks necessary for Protobuf management.
[EXTERNAL_DOWNLOADS]: The skill references the official buf.build service and its associated GitHub Actions (bufbuild/buf-setup-action). It also provides instructions for downloading the buf binary via Homebrew, Go, or direct releases from the official site for tool installation and configuration.

buf-cli-lint-breaking-build