The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external data in the form of existing repositories and codebases.
Ingestion points: Instructions in SKILL.md to read source code, inspect repository structures, and capture baseline behavior from local projects.
Boundary markers: The skill does not provide specific instructions to delimit or ignore instructions found within the analyzed codebase.
Capability inventory: The agent is tasked with reading files, writing a new project structure, and executing baseline commands from the legacy app.
Sanitization: No specific sanitization or filtering of the input code's content is requested.
[COMMAND_EXECUTION]: The skill instructions require the agent to execute commands from the existing application to establish a behavior baseline.
Evidence: SKILL.md includes a requirement to "Capture runnable baseline behavior when practical: commands, sample inputs, config examples, outputs, errors, and exit codes."
[EXTERNAL_DOWNLOADS]: The skill references several external Go libraries for use in the rewrite process.
Evidence: references/horizon-news-aggregation.md lists various libraries such as github.com/spf13/cobra, github.com/google/go-github, and github.com/yuin/goldmark. These are standard tools from well-known maintainers and organizations.

rewrite-in-go