The Agent Skills Directory

[METADATA_POISONING]: The skill metadata identifies the author as 'samber' and provides links to 'samber's official GitHub repositories, which directly contradicts the 'rockcookies' author context. This discrepancy suggests potential impersonation of a well-known open-source developer.
[INDIRECT_PROMPT_INJECTION]: The skill analyzes Go source files, creating a vulnerability surface where instructions hidden in code comments could influence the agent's behavior.
Ingestion points: The skill ingests untrusted content from all project files matching the **/*.go pattern.
Boundary markers: The instructions lack specific delimiters or directions for the agent to ignore or isolate instructions found within the source code being analyzed.
Capability inventory: The agent has permission to execute Bash commands for go, dlv, git, and golangci-lint.
Sanitization: There is no evidence of sanitization or validation of data extracted from project files before it is potentially passed to shell commands.
[EXTERNAL_DOWNLOADS]: The skill configuration and documentation specify the download of external tools from GitHub repositories.
Fetches the Delve debugger from github.com/go-delve/delve.
Fetches the errcheck utility from github.com/kisielk/errcheck.
[REMOTE_CODE_EXECUTION]: The skill uses go install to download and execute binaries from GitHub repositories at runtime.

golang-troubleshooting