The Agent Skills Directory

[SAFE]: The skill is designed for local filesystem management with a well-defined and restricted scope. It specifically includes a safety protocol to abort execution if the filesystem scope is set to sensitive areas like the OS root or user home directories.
[SAFE]: No network operations, external downloads, or remote code execution patterns were detected. All activities are confined to the local filesystem.
[SAFE]: The skill operates in a read-mostly mode for existing wiki content, only writing generated reports to a specific 'archive/' directory and appending to a local 'log.md' file, which aligns with its documented health-check purpose.
[INDIRECT_PROMPT_INJECTION]: The skill inherently processes untrusted data as it scans user-generated markdown files and 'Overview.md' for structural and conceptual issues.
Ingestion points: Reads all .md files recursively within the wiki root (Workflow Steps 1, 3, 6, 7).
Boundary markers: None identified; user-generated content is parsed directly for links and frontmatter.
Capability inventory: Limited to filesystem read and local file writes (archive/*.md, log.md). No network or shell execution capabilities are present.
Sanitization: No explicit sanitization or instruction-filtering of the ingested markdown content is performed before processing.

wiki-lint