skills/adamos486/skills/autobuild/Gen Agent Trust Hub

autobuild

Pass

Audited by Gen Agent Trust Hub on Aug 15, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes stack-specific quality gates (lint, test, typecheck) and generates a commits.sh script which is marked as executable using chmod +x in references/ORCHESTRATION.md.
  • [PROMPT_INJECTION]: The skill processes external implementation plans (plan-path) which serve as untrusted input. Malicious instructions or code within these plans could influence sub-agent behavior or lead to unintended command execution.
  • Ingestion points: The required <plan-path> document specified in SKILL.md.
  • Boundary markers: Lacking explicit guards against instructions embedded in the plan document itself within the sub-agent prompts in references/SUBAGENT-PROMPTS.md.
  • Capability inventory: Includes shell execution (Bash), file writing, and sub-agent creation across all reference files.
  • Sanitization: Relies on the TDD micro-structure but does not sanitize the natural language content of the plan or the interpolated variables.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 15, 2026, 03:55 AM
Security Audit — agent-trust-hub — autobuild