resume-cr

Pass

Audited by Gen Agent Trust Hub on Jun 28, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill uses npx clean-room-skill@latest to launch its task runner, which involves downloading the package from the npm registry if it is not locally available.
  • [COMMAND_EXECUTION]: The skill provides instructions to execute the clean-room-skill command-line interface to manage role-based task execution and environment configuration.
  • [PROMPT_INJECTION]: The skill defines a system for processing external task data through various JSON manifest files.
  • Ingestion points: Reads numerous files such as task-manifest.json and qc-report.json to determine the state and scope of work.
  • Boundary markers: Mandates validation against JSON schemas and SHA-256 hash checks to ensure artifact integrity and authenticity.
  • Capability inventory: Triggers the execution of the clean-room runner and updates unit statuses based on processed artifacts.
  • Sanitization: Implements strict data sanitization through schema verification and enforces separation between clean and contaminated data roots.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 28, 2026, 06:22 PM
Security Audit — agent-trust-hub — resume-cr