The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to build Go applications by processing user requirements and existing code, creating a surface where malicious instructions could be embedded in the input data.
Ingestion points: User requirements, code snippets, and architectural descriptions provided to the agent during the 'Analyze architecture' phase in SKILL.md.
Boundary markers: The skill does not define explicit delimiters or instructions to ignore embedded commands within user-supplied content.
Capability inventory: The skill allows the agent to generate complex Go code involving concurrency (in references/concurrency.md), network communication (gRPC, REST), and file system operations (via io interfaces in references/interfaces.md).
Sanitization: No sanitization or validation of user-provided input is specified in the skill's instructions.
[METADATA_POISONING]: The SKILL.md file contains metadata attributing the skill to a different author ('https://github.com/Jeffallan') than the expected vendor context. While common in shared templates, this represents a discrepancy in the skill's metadata consistency.

golang-pro