build-fix

Pass

Audited by Gen Agent Trust Hub on Mar 13, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill's primary function is to execute project-specific build and verification tools (e.g., tsc, mypy, cargo check, go vet). It is instructed to run these commands to collect errors and verify fixes. This creates a risk if the project environment or build configuration is malicious, as the skill will execute those defined commands.
  • [PROMPT_INJECTION]: The skill instructions contain a 'Guidance Alignment' section that directs the agent to 'Treat newer user task updates as local overrides for the active workflow branch,' which establishes a behavioral pattern of allowing instruction overrides during a session.
  • [PROMPT_INJECTION]: The skill possesses an attack surface for indirect prompt injection due to its core workflow of processing untrusted data.
  • Ingestion points: The agent reads external source code files and the stdout/stderr output from build tools (SKILL.md).
  • Boundary markers: The instructions do not specify the use of delimiters or 'ignore embedded instructions' warnings when reading external content.
  • Capability inventory: The skill has significant capabilities, including the ability to execute shell commands and modify file contents across the repository (SKILL.md).
  • Sanitization: There is no mention of sanitizing or validating error messages or code comments before they are processed by the agent's logic.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 13, 2026, 10:25 AM