The Agent Skills Directory

[SAFE]: The skill facilitates local wiki maintenance through static analysis of markdown files. It executes a bundled Python script that relies solely on standard library modules and performs no network operations or sensitive data access. The workflows involve standard file reading, reporting, and updating logs within the project scope.- [INDIRECT_PROMPT_INJECTION]: The skill processes user-controlled markdown files and incorporates filenames and link targets into reports presented to the agent. This represents an attack surface for indirect prompt injection, although the risk is minimal given the utility's context and focused capabilities.\n
Ingestion points: The lint-wiki.py script recursively reads all .md files within the provided vault directory ($CLAUDE_PROJECT_DIR).\n
Boundary markers: The script output is formatted as a plain text report without specific delimiters to isolate user-controlled strings (filenames/links) from the agent's instructions.\n
Capability inventory: The agent has the capability to write to Wiki/log.md, create new stub files for missing pages, and execute the local qmd CLI tool for search index updates.\n
Sanitization: No sanitization or escaping is performed on the data extracted from the wiki pages before it is included in the output report.

wiki-lint