init-deep

Pass

Audited by Gen Agent Trust Hub on Jun 1, 2026

Risk Level: SAFE
Full Analysis
  • [COMMAND_EXECUTION]: The skill uses common shell utilities including find, awk, wc, and sed to perform structural analysis and calculate metrics such as directory depth and line counts. These operations are restricted to the local environment and are used solely for project characterization.
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests data from the project environment to inform documentation generation. * Ingestion points: Reads project directory structure, existing AGENTS.md or CLAUDE.md files, and source code comments. * Boundary markers: None identified. * Capability inventory: Spawns exploration sub-agents and uses file-writing tools (Write, Edit) to create or update documentation. * Sanitization: None observed; the skill is designed to summarize and aggregate found project patterns into markdown format.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 1, 2026, 06:58 PM
Security Audit — agent-trust-hub — init-deep