syncing-claude-md

Pass

Audited by Gen Agent Trust Hub on Feb 25, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill operates on local files using basic system utilities like cat, grep, and diff. There are no external network requests, obfuscated code blocks, or suspicious administrative commands.- [PROMPT_INJECTION]: The skill processes content from .humaninloop/memory/constitution.md to update agent instructions in CLAUDE.md. While the intended purpose is synchronization, this creates a potential surface for indirect prompt injection if the source file is manipulated. Ingestion points: Reads from .humaninloop/memory/constitution.md. Boundary markers: No explicit delimiter or instruction-ignoring warnings are used during the text interpolation process. Capability inventory: The skill performs local file reads and identifies updates using system commands across references/SYNC-PATTERNS.md. Sanitization: There is no evidence of filtering or escaping logic applied to the content before it is moved between files.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 25, 2026, 02:12 AM