skills-auto-activation

Pass

Audited by Gen Agent Trust Hub on Mar 6, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill provides a complete Node.js script (skill-activator.js) intended to be installed and executed as a local Claude Code hook. This script runs in the user's local environment to process input and output JSON according to the hook protocol.
  • [PROMPT_INJECTION]: The core functionality of the skill is to programmatically inject "additionalContext" into the agent's prompt stream. This mechanism is used as a meta-instructional tool to guide the AI toward using relevant skills based on the content of the user's request, rather than attempting to bypass safety filters.
  • [DATA_EXPOSURE]: The provided hook script accesses the local file system to read configuration data from ~/.claude/skill-rules.json. It does not attempt to access sensitive files like credentials or SSH keys, nor does it perform any network operations to send data externally.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 6, 2026, 05:18 PM