safe-push

Pass

Audited by Gen Agent Trust Hub on Sep 17, 2026

Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes shell commands such as git, grep, xargs, and sleep. These are used to analyze diffs, scan tracked files, audit commit messages, and implement staggered push strategies for rate limiting.
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests data from external repository content and configuration files, which presents a potential injection surface.
  • Ingestion points: Content is retrieved from git diff, git log, and repository files via git ls-files. It also reads patterns from a user-managed configuration file at ~/.claude/safe-push-blocklist.
  • Boundary markers: No explicit boundary markers or 'ignore' instructions are used when processing repository content.
  • Capability inventory: The agent has the ability to execute shell commands and perform git push operations.
  • Sanitization: The skill filters comments and empty lines from the personal blocklist configuration but does not perform additional escaping of shell metacharacters before including the patterns in shell commands.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 17, 2026, 08:54 PM
Security Audit — agent-trust-hub — safe-push