The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses shell and git commands such as grep, git mv, and git rm to locate and rename placeholder strings across the project files and directories during the setup and self-cleanup phases.
[EXTERNAL_DOWNLOADS]: During the cleanup phase, the skill executes npx skills add padamson/ontology-authoring-template to fetch and install a tracked version of an external author-maintained skill.
[INDIRECT_PROMPT_INJECTION]: The skill has an attack surface for indirect prompt injection as it ingests and processes content from various local project files that could contain untrusted data. 1. Ingestion points: Reads local files including schema/*.yaml, data/*.yaml, *.toml, README.md, and CLAUDE.md. 2. Boundary markers: No boundary markers or instructions to ignore embedded content are implemented. 3. Capability inventory: The skill can move and delete files, modify their contents, and execute shell scripts and external tool installers. 4. Sanitization: No sanitization or validation of the ingested text is performed during the global renaming process.
[DYNAMIC_EXECUTION]: The skill modifies configuration files and potentially script-related paths that are subsequently used by other tools (e.g., mdbook build) and executes a local installation script (./scripts/install-assets.sh).

setup-ontology