opsx-explore

Pass

Audited by Gen Agent Trust Hub on Aug 14, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes local command-line utilities openspec and codegraph to list project changes and explore codebase symbols.
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from the filesystem and user input, creating a surface for embedded instructions.
  • Ingestion points: Arguments provided to the /opsx:explore command and contents of change artifacts or source files read during investigation.
  • Boundary markers: None present; the instructions lack delimiters or explicit warnings to ignore embedded directives in external files.
  • Capability inventory: Includes command execution and file reading, but strictly forbids writing code or implementing features.
  • Sanitization: No mechanisms are specified for sanitizing or validating the ingested content.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 14, 2026, 05:53 PM
Security Audit — agent-trust-hub — opsx-explore