skills/marcoax/skills/autoresearch/Gen Agent Trust Hub

autoresearch

Pass

Audited by Gen Agent Trust Hub on Apr 23, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [COMMAND_EXECUTION]: The skill uses the shell command open to automatically display the generated HTML dashboard in the default browser on macOS.
  • [PROMPT_INJECTION]: The skill exposes an indirect prompt injection surface by ingesting and processing content from external skill files and their generated outputs.
  • Ingestion points: Target skill content and runtime output data (SKILL.md).
  • Boundary markers: No specific delimiters are defined to separate ingested data from the agent's core optimization instructions.
  • Capability inventory: The skill utilizes file system access (read/write), terminal command execution (open), and autonomous recursive execution cycles (SKILL.md).
  • Sanitization: No sanitization or filtering is applied to external data processed during the optimization loop.
  • [EXTERNAL_DOWNLOADS]: The generated dashboard includes a reference to the Chart.js library via a public CDN to facilitate browser-based performance visualization.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 23, 2026, 09:48 AM
Security Audit — agent-trust-hub — autoresearch