The Agent Skills Directory

[SAFE]: No security concerns were identified. The skill acts as an infrastructure specialist, gathering information through standard prompts and producing documentation-based outputs.
[INDIRECT_PROMPT_INJECTION]: The skill's primary function involves processing user-provided infrastructure details, which constitutes an ingestion surface for indirect prompt injection. However, given the purely instructional and advisory nature of the skill, the risk is minimal.
Ingestion points: Context gathering in 'Step 1' of SKILL.md.
Boundary markers: Not explicitly defined in the prompt instructions.
Capability inventory: Tools listed in frontmatter include Bash, Write, WebFetch, and WebSearch.
Sanitization: No explicit validation or filtering of user-supplied context is performed.
[DATA_EXPOSURE]: No hardcoded credentials, sensitive file paths, or unauthorized data access patterns were found. The skill requests context about the user's setup as part of its legitimate design workflow.
[REMOTE_CODE_EXECUTION]: No patterns for downloading and executing remote scripts or dynamic code generation from untrusted sources were detected.

terra-drift