The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process Go source code (**/*.go) and has the ability to perform file modifications and execute development tools via the shell (go, git, golangci-lint). This creates a surface where an attacker could embed malicious instructions in code comments (e.g., inside a pull request being reviewed) that the agent might inadvertently follow.
Ingestion points: The skill targets all .go files within the workspace.
Boundary markers: There are no explicit instructions or delimiters provided to help the agent distinguish between code/comments and instructions, nor are there warnings to ignore embedded commands.
Capability inventory: The skill allows the use of Edit, Write, and Bash (for go, git, and golangci-lint), which provide sufficient privilege to carry out actions if the agent is successfully injected.
Sanitization: The skill lacks mechanisms to sanitize or filter out natural language instructions found within the code data it processes.

golang-naming