glossary-management

Pass

Audited by Gen Agent Trust Hub on Nov 7, 2026

Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill uses the Bash tool to execute local TypeScript scripts via the bun runtime for synchronization and validation tasks.
  • Evidence: Commands such as bun ~/.claude/tools/bin/glossary-sync.ts and bun ~/eon/cc-skills/plugins/itp-hooks/hooks/posttooluse-terminology-sync.ts in SKILL.md.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes data from multiple project-level markdown files, creating an attack surface where embedded instructions could influence agent behavior.
  • Ingestion points: Reads from project CLAUDE.md files found in directories specified by the SCAN_PATHS configuration (e.g., ~/eon/*/CLAUDE.md).
  • Boundary markers: Relies on a specific 5-column markdown table schema for identifying and extracting terminology data.
  • Capability inventory: The skill utilizes Bash, Read, Edit, and Grep tools, which allow it to modify local files and execute maintenance scripts.
  • Sanitization: No explicit logic or instructions are provided to sanitize, escape, or ignore natural language instructions that might be embedded within the ingested markdown files.
Audit Metadata
Risk Level
SAFE
Analyzed
Nov 7, 2026, 02:12 AM