The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes local TypeScript scripts using the Bun runtime, specifically ~/.claude/tools/bin/glossary-sync.ts and ~/eon/cc-skills/plugins/itp-hooks/hooks/posttooluse-terminology-sync.ts. These scripts are used to synchronize terminology and are part of the intended local developer environment setup.
[COMMAND_EXECUTION]: Uses standard Unix utilities including grep for format validation, find for file discovery, and rm for cleaning up stale directory structures.
[INDIRECT_PROMPT_INJECTION]: The skill processes content from various CLAUDE.md files and a central GLOSSARY.md file. This ingestion of external data constitutes a potential attack surface if project files contain malicious content designed to exploit the sync scripts.
Ingestion points: Files located via SCAN_PATHS (e.g., ~/eon/*/CLAUDE.md) and the central ~/.claude/docs/GLOSSARY.md file.
Boundary markers: The skill enforces a strict 5-column markdown table schema and utilizes specific HTML comments () for configuration.
Capability inventory: The skill has permissions for Bash execution, Read/Edit file access, and Grep search operations.
Sanitization: Validation is performed through regular expression checks (via grep) and structured parsing within the local TypeScript hooks.

glossary-management