feature-audit

Pass

Audited by Gen Agent Trust Hub on May 15, 2026

Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [PROMPT_INJECTION]: The skill processes untrusted content from the repository (code, tests, documentation) as part of its audit workflow, creating an indirect prompt injection surface.
  • Ingestion points: The "Discovery Workflow" and "Map the feature surface" sections in SKILL.md instruct the agent to read project manifests, config files, source code, and documentation from the audited repository.
  • Boundary markers: No explicit instructions to treat code comments or documentation as untrusted data or to use specific delimiters (e.g., XML tags or markdown blocks) to isolate untrusted content are provided.
  • Capability inventory: The agent is authorized to run terminal commands (lint, test runners) and, if requested by the user in the "Post-Audit Fix Workflow", modify files to apply fixes.
  • Sanitization: No explicit sanitization, validation, or escaping of the ingested repository content is mentioned before it is processed or used to generate audit reports and fix recommendations.
  • [COMMAND_EXECUTION]: The skill executes shell commands based on existing project scripts to verify code quality and functionality.
  • Evidence: The "Verify safely" section in SKILL.md directs the agent to run linting, typechecking, and unit tests using project-defined scripts found in manifests or documentation.
Audit Metadata
Risk Level
SAFE
Analyzed
May 15, 2026, 05:33 PM
Security Audit — agent-trust-hub — feature-audit