skills/vllnt/skills/improve-code/Gen Agent Trust Hub

improve-code

Pass

Audited by Gen Agent Trust Hub on Nov 6, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONNO_CODE
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and analyze external codebases, including comments and configuration files, which are untrusted inputs. This provides a potential attack surface for indirect prompt injection.
  • Ingestion points: Processes external source code, comments, and project configuration (SKILL.md, workspace-turborepo/REFERENCE.md).
  • Boundary markers: Lacks explicit delimiters or specific instructions for the agent to ignore embedded commands within the analyzed content.
  • Capability inventory: Possesses the ability to modify source code, execute test suites, and adjust build/CI configurations (SKILL.md, ci-optimization/REFERENCE.md).
  • Sanitization: Relies on a multi-stage validation protocol (quality-validation.md) involving self-assessment and peer review rather than automated sanitization.
  • [NO_CODE]: The skill consists solely of Markdown documentation and does not include any scripts, binaries, or automated installation commands.
Audit Metadata
Risk Level
SAFE
Analyzed
Nov 6, 2026, 04:54 PM