The Agent Skills Directory

[SAFE]: The skill provides comprehensive and normative guidance for Go naming conventions, aligning with standard industry practices such as Effective Go and the Uber Go Style Guide.
[COMMAND_EXECUTION]: The skill includes a local utility script (scripts/check-naming.sh) that identifies style violations in Go source files. The script uses standard Unix utilities like find and sed for pattern matching. It does not perform network operations, access sensitive user credentials, or modify system configurations.
[INDIRECT_PROMPT_INJECTION]: The skill has a surface for indirect prompt injection as it processes external Go source code which could contain malicious comments or identifiers designed to influence the agent's behavior during a refactoring task.
Ingestion points: Go source files located in the directory provided to scripts/check-naming.sh or the agent's context.
Boundary markers: None explicitly implemented in the provided script, though the script performs specific regex-based extraction rather than open-ended analysis.
Capability inventory: Execution of the Bash linter script and file reading via standard tools.
Sanitization: The script uses restricted regular expressions to match specific naming patterns, which limits the impact of arbitrary content in the files.

go-naming