The Agent Skills Directory

[SAFE]: The skill uses the uv package manager for dependency handling. While pyyaml is declared as a dependency in scripts/vault_search.py, it is currently unused in the script's logic, and no other external libraries are required.
[COMMAND_EXECUTION]: The script scripts/vault_search.py executes the ripgrep utility to search Markdown files. This execution is performed using subprocess.run with arguments passed as a list and shell=False, which prevents shell command injection vulnerabilities even when handling natural language queries.
[PROMPT_INJECTION]: The skill processes content from local Markdown files, which represents an indirect prompt injection surface.
Ingestion points: Markdown files in the vault accessed via ripgrep output.
Boundary markers: None explicitly defined in the script; the AI agent is responsible for orchestrating the tool and interpreting the output.
Capability inventory: Local file search and metadata retrieval across .md files.
Sanitization: Individual search terms are escaped using re.escape() before being used in search patterns.

obsidian-vault-search