opencode-command-authoring

Pass

Audited by Gen Agent Trust Hub on Sep 18, 2026

Risk Level: SAFECOMMAND_EXECUTIONDYNAMIC_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill utilizes local binaries bin/create-command and bin/opencode-command for command lifecycle management. These tools appear to be vendor-provided resources associated with the skill author.
  • [DYNAMIC_EXECUTION]: The skill's primary function is the dynamic generation and execution of command files located in the .opencode/command/ directory.
  • [INDIRECT_PROMPT_INJECTION]: The skill accepts user-supplied command names, purposes, and descriptions to generate markdown files that are later executed, presenting a surface for instruction injection.
  • Ingestion points: Command name, purpose, and frontmatter description inputs.
  • Boundary markers: The instructions lack explicit delimiters or warnings to isolate user input from the generated command's executable context.
  • Capability inventory: Local shell command execution and file system write operations to the .opencode/ directory.
  • Sanitization: No mechanisms for sanitizing or validating user input are identified in the skill documentation.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 18, 2026, 10:49 AM
Security Audit — agent-trust-hub — opencode-command-authoring