beyond-code-verify

Pass

Audited by Gen Agent Trust Hub on Aug 29, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill dynamically identifies and executes project-specific tooling commands for linting, typechecking, building, and running tests. These commands are derived from the project environment.
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data in the form of git diff outputs and project documentation (spec.md, plan.md) to be processed by a subagent for auditing. This represents a surface where malicious instructions could be embedded in code comments or requirements documents to influence the agent's behavior.
  • Ingestion points: Reads git diff, .beyond-code/<slug>/spec.md, and .beyond-code/<slug>/plan.md in Stage 1.
  • Boundary markers: None identified. The content is passed to the subagent as part of the audit prompt.
  • Capability inventory: The skill has the ability to execute shell commands (lint, test, ls) and perform file system operations (moving directories for archiving).
  • Sanitization: No escaping or validation is performed on the ingested diff or documentation content before it is processed.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 29, 2026, 04:32 PM
Security Audit — agent-trust-hub — beyond-code-verify