rename-references
Pass
Audited by Gen Agent Trust Hub on Jul 30, 2026
Risk Level: SAFE
Full Analysis
- [COMMAND_EXECUTION]: The skill uses dynamic context injection via the
!commandsyntax to execute benign, read-only git commands (git status,git diff,git branch) at load time. These commands are used to establish repository state and identify recent renames without incorporating untrusted user input into the shell execution. - [DATA_EXFILTRATION]: The skill scans project files for identifier references. It includes clear logic to auto-exclude sensitive directories, such as the
~/.claude/projects/memory paths, by default. There are no network-based exfiltration patterns or external communication tools configured. - [PROMPT_INJECTION]: The skill manages the risk of indirect prompt injection—where instructions might be hidden in the code being scanned—by implementing a mandatory triage and confirmation workflow. Matches found in prose or containing common English verbs (the 'Ambiguous' bucket) are never auto-applied and must be reviewed by the user individually.
- [SAFE]: The skill follows established safety best practices, including idempotency in its edit operations and clear boundaries against out-of-scope tasks like AST-level refactoring or framework migrations. No malicious obfuscation, persistence, or privilege escalation patterns were detected.
Audit Metadata