config-export

Pass

Audited by Gen Agent Trust Hub on Aug 14, 2026

Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill uses a bash script (core/scripts/config-export.sh) for deterministic delegation of agent and command renders. This is a local process within the skill's infrastructure used to maintain parity between different tool outputs.
  • [REMOTE_CODE_EXECUTION]: No remote code execution patterns were detected. The transformation process relies on in-context LLM reasoning and local scripts.
  • [DATA_EXFILTRATION]: The skill contains logic specifically designed to strip sensitive Claude Code execution tokens (like mem_save, mem_search, or ~/.claude/ paths) before writing configuration to other tools, reducing the risk of accidental data exposure.
  • [PROMPT_INJECTION]: The skill uses 'STRIP' guards and preambles to prevent internal agent instructions or slash commands from leaking into the exported instruction files for other AI assistants.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 14, 2026, 09:46 AM
Security Audit — agent-trust-hub — config-export