openspec-explore

Pass

Audited by Gen Agent Trust Hub on Jul 29, 2026

Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill uses the openspec CLI (specifically openspec list --json) to retrieve project state. This command is restricted to the skill's own tooling and does not involve arbitrary shell execution.
  • [PROMPT_INJECTION]: The skill processes codebase files and OpenSpec artifacts, creating a surface for indirect prompt injection.
  • Ingestion points: Files located in openspec/changes/ and the codebase generally (e.g., proposal.md, design.md, tasks.md).
  • Boundary markers: The instructions do not define specific delimiters or instructions to ignore embedded commands for external content.
  • Capability inventory: The agent can read files and write OpenSpec artifacts. The skill contains a strong negative constraint against implementing application code or features.
  • Sanitization: No specific sanitization or validation mechanisms for ingested file content are described.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 29, 2026, 12:04 PM
Security Audit — agent-trust-hub — openspec-explore