shipping-code

Pass

Audited by Gen Agent Trust Hub on Jun 22, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill invokes several local command-line tools for build verification and version control, including git, cargo (for Rust projects), npx (executing eslint and tsc for TypeScript projects), and a project-specific tool named trader for migration verification.
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it processes data from git diff, which represents an untrusted ingestion point.
  • Ingestion points: Staged code changes are read via git diff --cached commands in SKILL.md.
  • Boundary markers: None identified. The instructions do not specify delimiters or provide warnings to the agent to ignore instructions embedded within the code being analyzed.
  • Capability inventory: The skill has the capability to execute shell commands such as git commit, cargo build (which can run arbitrary code via build scripts), and npx commands.
  • Sanitization: No sanitization or filtering is applied to the content of the diffs before they are processed by the agent.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 22, 2026, 09:49 AM
Security Audit — agent-trust-hub — shipping-code